Spatial Configuration and Three-Dimensional Conformation Directed Design, Synthesis, Antiviral Activity, and Structure-Activity Relationships of Phenanthroindolizidine Analogues

Article abstract of DOI:10.1021/acs.jafc.5b06112

Our recent investigation on the antiviral activities against tobacco mosaic virus (TMV) of phenanthroindolizidine alkaloid analogues preliminarily revealed that the basic skeleton and substitution pattern at the C13a position of the molecule, which are cl

Full text of DOI:10.1021/acs.jafc.5b06112

Article
pubs.acs.org/JAFC
Spatial Configuration and Three-Dimensional Conformation Directed
Design, Synthesis, Antiviral Activity, and Structure−Activity
Relationships of Phenanthroindolizidine Analogues
Bo Su, Chunlong Cai, Meng Deng, and Qingmin Wang*
State Key Laboratory of Elemento-Organic Chemistry, Research Institute of Elemento-Organic Chemistry, Collaborative Innovation
Center of Chemical Science and Engineering (Tianjin), Nankai University, Tianjin 300071, People’s Republic of China
S
* Supporting Information
ABSTRACT: Our recent investigation on the antiviral activities against tobacco mosaic virus (TMV) of phenanthroindolizidine
alkaloid analogues preliminarily revealed that the basic skeleton and substitution pattern at the C13a position of the molecule,
which are closely related to the spatial arrangement of the molecule, have great effects on the biological activity. To further study
the in-depth influence of spatial configuration and three-dimensional (3D) conformation of the molecules on their anti-TMV
activities and related structure−activity relationship (SAR), a series of D-ring opened derivatives 3, 4, 5a−5j, 6, and 7, chiral 13a-
and/or 14-substituted phenanthroindolizidine analogues 10−12 and 18−20, and their enantiomers ent-10−ent-12 and ent-18−
ent-20 were synthesized and evaluated for their anti-TMV activities. Bioassay results showed that most of the chiral
phenanthroindolizidines displayed good to excellent in vivo anti-TMV activity. Among these compounds, ent-11 showed more
potent activity than Ningnanmycin (one of the most successful commercial antiviral agents), thus emerging as a potential
inhibitor of the plant virus. Further SARs were also discussed for the first time under the chiral scenario, demonstrating that both
spatial configuration and 3D conformation of the molecules are crucial for keeping high anti-TMV activity.
KEYWORDS: phenanthroindolizidine alkaloid, antiviral activity, tobacco mosaic virus, structure−activity relationship, anti-TMV
INTRODUCTION
■
Tabacco mosaic virus (TMV), the first virus that had ever been
discovered and purified, has been known for over 1 century,
dating back to Benjerinck’s description of mosaic disease of
tobacco as a contagium vivum fluidum and the modern usage of
the term “virus”.1 It is known that more than 125 individual
species of 9 plant families (such as tobacco, pepper, cucumber,
tomato, many ornamental flowers, etc.) can be infected by
TMV.2 TMV is the most persistent plant virus known, which
can survive in dry plant debris for as long as 100 years;
therefore, it is very difficult to control once plants have been
infected. TMV causes serious damage worldwide, and the
economic loss each year is up to U.S. $100 million.3
Ningnanmycin (Figure 1), displaying 56.0% inhibitory effect
at 500 μg mL−1, is almost the most effective commercially
available anti-plant virus agent. Ribavirin (Figure 1), another
plant virus inhibitor, is widely used, and its inhibitory effect is
always less than 50% at 500 μg mL−1. Because TMV causes
large economic loss every year and commercial plant virucides
could always not show a satisfactory inhibitory effect (usually
30−60%), the development of novel and more potent antiplant
virus agents is very appealing. Consequently, as a result of their
known antiviral activities, some chemicals, such as pyrazole
derivatives,4 nucleotides,5 α-aminophosphonate derivatives,6 3-
acetonyl-3-hydroxyoxindole,7 triazolyl compounds,8 oxidized
polyamines,9 and substituted phenylureas,10 have been
investigated extensively. However, few of them have been
successfully developed and applied in agriculture.
Figure 1. Chemical structures of Ningnanmycin, ribavirin, and
representative phenanthroindolizidine alkaloids.
decompose, and specific to a target species, have a unique mode
of action, etc.11,12 Phenanthroindolizidine alkaloids are a small
class of pentacyclic natural products,13 which possess a range of
biological activities, such as anticancer, anti-inflammatory,
antiamoebic, antilupus effects, etc.14−18 Ten years ago, our
group initiated a project aiming at seeking natural biologically
active compounds as alternatives to synthetic chemicals and
first found that the alcohol extract of Cynanchum komarovii
showed a moderate inhibitory effect against TMV.19 (R)-
Received: December 26, 2015
Revised: February 17, 2016
Accepted: February 28, 2016
Published: February 28, 2016
In comparison to synthetic chemicals, natural-product-based
TMV inhibitors are less toxic to the environment, easier to
© 2016 American Chemical Society
2039
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045

Journal of Agricultural and Food Chemistry
Article
Figure 2. Design of the targeted molecules based on spatial configuration and 3D conformation and modification.
Antofine was then proven to be the main active compound
(Figure 1). Mechanism studies against TMV indicated that, as a
result of the favorable interaction with the origin of TMV RNA
(oriRNA), antofine exhibits its antiviral activity by binding to
oriRNA and interfering with virus assembly initiation.20
Distinguished from the common 13a-H phenanthroindolizidine
alkaloids, hypoestestatin 1 (Figure 1) possesses a methyl group
at the C13a position.21 Since its discovery in 1984, several other
13a-methyl-phenanthroindolizidine alkaloids were subsequently
reported.22,23 However, in comparison to their 13a-H counter-
parts, these structurally novel 13a-substituted antofine ana-
logues were rarely investigated.
To further systematically investigate the in-depth influence of
spatial configuration and 3D conformation of phenanthroindo-
lizidine alkaloids on the antiviral activities (Figure 2), a series of
D-ring opened 13a-substituted alkaloid analouges and chiral
13a- and/or 14-substituted alkaloids were designed, synthe-
sized, and evaluated for their antiviral activity against TMV.
The SAR study of these compounds was also discussed for the
first time under the chiral scenario.
MATERIALS AND METHODS
■
Chemicals. Reagents, such as proline, di-tert-butyl dicarbonate,
trifluoroacetyl chloride, acetyl chloride, trimethylacetyl chloride,
benzoyl chloride, ethyl carbonochloridate, benzyl carbonochloridate,
benzyl carbonochloridate, 4-methylbenzene-1-sulfonyl chloride, meth-
yl iodide, benzyl bromide, thionyl chloride, bromine, triethylamine,
and potassium carbonate, were purchased from Tianjin Reagents 6th
Company and used directly without further purification. Sodium, tert-
butryl lithium, lithium diisopropylamide, lithium aluminum hydride,
and lithium triethylborohydride were purchased from Alfa Aesar and
used as received. All anhydrous solvents, such as tetrahydrofuran
(THF), dichloromethane (DCM), and dimethylformamide (DMF),
purchased from Tianjin Reagents 6th Company, were dried and
purified according to standard techniques just before use. Other
solvents, such as methanol, ethanol, petroleum ether, and ethyl acetate,
were purchased from Tianjin Reagents 6th Company and used directly
without further purification.
Although a structure−activity relationship (SAR) study on
13a-H antofine analogues has been investigated exten-
sively,24−28 the previous research was only limited to
substitution effects on the phenanthryl ring and precursors en
route to these natural products or analogues, mainly as a result
of the underdevelopment of the synthetic methodology about
these alkaloids. However, a simple modification of the
phenanthryl ring of the pentacyclic structure of phenanthroin-
dolizidine alkaloid can only adjust its electronic or steric
property but does not have much influence on its spatial
configuration and three-dimensional (3D) conformation. In
fact, the latter factors play a more important role in adjusting
the delicate interaction between the specific molecule and the
target and, thus, are more essential for the investigation of in-
depth SARs. Toward this goal, we initially designed and
synthesized two types of structurally novel antofine analogues,
which changed the size of the D ring or the arrangement of D/
E rings from the mother pentacyclic phenanthroindolizidine
significantly. Bioassay results demonstrated for the first time
that the original spatial arrangement of phenanthroindolizidines
was not optimal for these alkaloids to keep high activities.29
Inspired by the unique structure of the 13a-methyl-phenan-
throindolizidine alkaloid hypoestestatins 1 and 2 (Figure 1), we
then introduced a series of functional groups at the bridgehead
of the D/E ring of phenanthroindolizidine alkaloids, largely
varying their 3D conformation from their 13a-H counterparts.
Antivirus activity results showed that some of the C-13a-
substituted derivatives were indeed much more potent than
antofine.30 Therefore, these results demonstrated that further
studies based on this line was in demand.
Instruments. High-resolution mass spectra (HRMS) were
recorded on Fourier transform ion cyclotron resonance mass
spectrometry (FT-ICR MS, Ionspec, 7.0 T). Melting points were
determined on an X-4 binocular microscope melting point apparatus
and were uncorrected.
1-tert-Butyl-2-methyl-2-((3,6,7-trimethoxyphenanthren-9-
yl)methyl)pyrrolidine-1,2-dicarboxylate (3). To a solution of i-
Pr2NH (0.9 g, 9.0 mmol) in THF (10 mL) was added n-BuLi (4.3 mL,
2.2 M in hexane, 9.5 mL) dropwise at −78 °C under Ar. After 10 min,
a solution of compound 2 (2.06 g, 9.0 mmol) in THF (10 mL) was
added dropwise. The reaction was warmed to −30 °C and stirred at
this temperature for 30 min, and then a solution of compound 1 (2.16
g, 6.0 mmol) in THF (100 mL) was added. The reaction was stirred at
room temperature for 3 h and then was quenched with aqueous
saturated ammonium chloride (100 mL). After separation, the aqueous
phase was extracted with ethyl acetate (EA) (100 mL × 2). Combined
organic layers were evaporated in vacuo, and the residue was purified
by column chromatography on silica gel, giving compound 3 (2.8 g,
92%) as a white solid, with a melting point (mp) of 73−75 °C.
Because of the existence of two kinetically distinct conformations of
compound 3, two sets of nuclear magnetic resonance (NMR) were
2040
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045

Journal of Agricultural and Food Chemistry
Article
Figure 3. Synthetic route of D-ring opened phenanthroindolizidine alkaloid analogues.
observed. For NMR data, see the Supporting Information. HRMS
calcd for C29H35NO7 (M + Na)+, 532.2306; found, 532.2310.
Methyl-2-((3,6,7-trimethoxyphenanthren-9-yl)methyl)-
pyrrolidine-2-carboxylate (4). To a solution of compound 3 (4.0 g,
7.86 mmol) in MeOH (200 mL) was added aqueous HCl (12 M, 3
mL). The reaction was heated at reflux for 2 h, and then the solvent
was evaporated in vacuo. To the residue was added water (200 mL)
and aqueous NaOH to pH 14, and the resulting solution was extracted
with DCM (80 mL × 3). Combined organic layers were washed with
brine, dried over Na2SO4, filtered, and evaporated to give compound 4
(3.2 g, 94%) as a white solid, with a mp of 162−163 °C. For NMR
data, see the Supporting Information. HRMS [matrix-assisted laser
desorption/ionization (MALDI)] m/z calcd for C24H28NO5 (M +
H)+, 410.1962; found, 410.1965.
Methyl-1-(2,2,2-trifluoroacetyl)-2-((3,6,7-trimethoxyphe-
nanthren-9-yl)methyl)pyrrolidine-2-carboxylate (5a). To a
solution of compound 4 (0.41 g, 1.0 mmol) and 4-dimethylaminopyr-
idine (DMAP) (0.15 g, 1.2 mmol) in DCM (40 mL) was added a
solution of trifluoroacetic anhydride (TFAA) (0.20 g, 1.1 mmol) in
DCM (10 mL) dropwise at room temperature. After 2 h, the reaction
mixture was washed with diluted aqueous HCl, water, and brine. After
the solution was dried over Na2SO4, it was evaporated and the
resulting residue was purified by column chromatography on silica gel
to give compound 5a (0.46 g, 92%) as a white solid, with a mp of
174−175 °C. For NMR data, see the Supporting Information. HRMS
(MALDI) m/z calcd for C26H26F3NO6Na (M + Na)+, 528.1604;
found, 528.1603.
Methyl-1-acetyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-2-carboxylate (5b). Compound 5b was
obtained using a synthetic procedure similar to that of compound
5a as a white solid (85%), with a mp of 208−210 °C. For NMR data,
see the Supporting Information. HRMS (MALDI) m/z calcd for
C26H29NO6Na (M + Na)+, 474.1887; found, 474.1895.
Methyl-1-pivaloyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-2-carboxylate (5c). Compound 5c was
obtained using a synthetic procedure similar to that of compound
5a as a white solid (92%), with a mp of 188−189 °C. For NMR data,
see the Supporting Information. HRMS (MALDI) m/z calcd for
C29H35NO6Na (M + Na)+, 516.2357; found, 516.2350.
1-Benzyl-2-methyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-1,2-dicarboxylate (5f). Compound 5f was
obtained using a synthetic procedure similar to that of compound
5a as a white solid (90%), with a mp of 153−155 °C. For NMR data,
see the Supporting Information. HRMS (MALDI) m/z calcd for
C32H33NO7Na (M + Na)+, 566.2149; found, 566.2149.
Methyl-1-(methylsulfonyl)-2-((3,6,7-trimethoxyphenanth-
ren-9-yl)methyl)pyrrolidine-2-carboxylate (5g). Compound 5g
was obtained using a synthetic procedure similar to that of compound
5a as a white solid (88%), with a mp of 169−171 °C. For NMR data,
see the Supporting Information. HRMS (MALDI) m/z calcd for
C25H29NO7SNa (M + Na)+, 510.1557; found, 510.1555.
Methyl-1-tosyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-2-carboxylate (5h). Compound 5h was
obtained using a synthetic procedure similar to that of compound
5a as a white solid (88%), with a mp of 205−206 °C. For NMR data,
see the Supporting Information. HRMS (MALDI) m/z calcd for
C31H33NO7SNa (M + Na)+, 586.1870; found, 586.1863.
Methyl-1-methyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-2-carboxylate (5i). To a solution of compound
4 (0.37 g, 0.91 mmol) and aqueous formic aldehyde (30%, 0.15 mL)
in MeOH (40 mL) was added NaBH3CN (0.12 g, 1.9 mmol). The
reaction was stirred at room temperature for 4 h and then quenched
with saturated aqueous amonium chloride (50 mL). The solution was
extracted with DCM (30 mL × 3), and the combined organic layers
were washed with brine, dried over Na2SO4, filtered, and evaporated.
The resulting residue was purified by column chromatography on silica
gel to give compound 5i (0.37 g, 94%) as a white solid, with a mp of
108−110 °C. For NMR data, see the Supporting Information. HRMS
(MALDI) m/z calcd for C25H29NO5Na (M + Na)+, 446.1937; found,
446.1941.
Methyl-1-benzyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-2-carboxylate (5j). Compound 5j was ob-
tained using a synthetic procedure similar to that of compound 5i as a
white solid (89%), with a mp of 110−112 °C. For NMR data, see the
Supporting Information. HRMS (MALDI) m/z calcd for
C31H33NO5Na (M + Na)+, 522.2251; found, 522.2256.
tert-Butyl-2-(hydroxymethyl)-2-((3,6,7-trimethoxyphenanth-
ren-9-yl)methyl)pyrrolidine-1-carboxylate (6). To a suspension
of LiAlH4 (85 mg, 2.2 mmol) in THF (100 mL) at −78 °C was added
compound 3 (0.94 g, 1.85 mmol) in one portion. The reaction was
stirred at −20 °C for 2 h and then quenched with saturated aqueous
ammonium chloride. The solvent was evaporated in vacuo, and the
resulting residue was dissolved in DCM, which was washed with water
and brine. After the solution was dried over Na2SO4, filtered, and
evaporated, the resulting residue was purified by column chromatog-
raphy on silica gel to give compound 6 (0.81 g, 92%), with a mp of
178−180 °C. For NMR data, see the Supporting Information. HRMS
(MALDI) m/z calcd for C28H35NO6Na (M + Na)+, 504.2357; found,
504.2359.
Methyl-1-benzoyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-2-carboxylate (5d). Compound 5d was
obtained using a synthetic procedure similar to that of compound
5a as a white solid (90%), with a mp of 204−206 °C. For NMR data,
see the Supporting Information. HRMS (MALDI) m/z calcd for
C31H31NO6Na (M + Na)+, 536.2044; found, 536.2042.
1-Ethyl-2-methyl-2-((3,6,7-trimethoxyphenanthren-9-yl)-
methyl)pyrrolidine-1,2-dicarboxylate (5e). Compound 5e was
obtained using a synthetic procedure similar to that of compound 5a
as a white solid (92%), with a mp of 129−131 °C. For NMR data, see
the Supporting Information. HRMS (MALDI) m/z calcd for
C27H31NO7Na (M + Na)+, 504.1993; found, 504.1996.
tert-Butyl-2-(methoxymethyl)-2-((3,6,7-trimethoxyphe-
nanthren-9-yl)methyl)pyrrolidine-1-carboxylate (7). To a sol-
2041
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045

Journal of Agricultural and Food Chemistry
Article
Figure 4. Enantioselective synthetic route of 13a-substituted phenanthroindolizidine alkaloid analogues.
Figure 5. Enantioselective synthetic route of optical 13a-methyl-14-substituted phenanthroindolizidine alkaloid analogues.
ution of compound 6 (0.47 g, 0.98 mmol) and BF3·Et2O (0.01 mL) in
DCM (20 mL) was added a solution of CH2N2 in ether until the
starting material was comsumed completely, which was monited by
thin-layer chromatography (TLC). After the reaction was completed,
the reaction was quenched with acetic acid. The reaction mixture was
then washed with water and brine, dried over Na2SO4, filtered, and
evaporated, and the resulting residue was purified by column
chromatography on silica gel to give compound 7 (0.36 g, 73%),
with a mp of 129−130 °C. For NMR data, see the Supporting
Information. HRMS (MALDI) m/z calcd for C29H37NO6Na (M +
Na)+, 518.2513; found, 518.2507.
RESULTS AND DISCUSSION
■
Chemistry. As shown in Figure 3, the synthesis of the D-
ring opened phenanthroindolizidine alkaloid analogues com-
menced with the coupling of two known compounds,
phenanthryl bromide 1 and proline derivative 2,33 giving
compound 3 in 92% yield. Under acidic conditions, the
protecting group (Boc) of compound 3 was easily removed,
resulting in amine 4 in 94% yield. Compounds 5a−5j were
readily prepared from compound 4 via corresponding acylation,
sulfonylation, and alkylation reactions. Treatment of compound
3 with LiAlH4 gave reduced compound 6, which was then
methylated using diazomethane as a methylation reagent. The
enantioselective synthesis of 13a-substituted phenanthroindoli-
zidine alkaloid analogues was shown in Figure 4.31 Using
phenanthryl bromide 1 and optical oxazolidinone 8 as starting
material, enantiopure compound 9 could be obtained as a single
diastereoisomer in good yield. Compound 9 was then
transformed to 13a-methylcarboxylate phenanthroindolizidine
10 via a sequential base-promoted ring opening of
oxazolidinone, followed by acid-promoted deprotection of the
2,2-disubstituted proline derivative and Pictet−Spengler
cyclization. Ester 10 was reduced by LiAlH4 in a nearly
quantitative yield to give alcohol 11. After extensive screening
The other compounds used for the bioassay were prepared
according to our recently developed synthetic methods.31,32
Antiviral Biological Assay.24 The bioassays of in vitro and in vivo
anti-TMV activity were carried out on Nicotiana tabacum L. The upper
leaves of N. tabacum L. were first inoculated with TMV, then were
selected and ground in phosphate buffer, and then filtered through
double-layer pledget. The filtrate was centrifuged, treated with
polyethylene glycol (PEG) twice, and then centrifuged again. The
whole experiment was processed at 4 °C. The main difference of three
in vivo modes was as follows: For protective effects, the leaves were
first treated with a solution of specific compound and then were
inoculated with TMV. For the inactivation effect, the leave was directly
inoculated with TMV that has been treated with a solution of specific
compound. For the curative effect, the leaves were first inoculated with
TMV and then treated with a solution of specific compound.
2042
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045

Journal of Agricultural and Food Chemistry
Article
of reaction conditions, the hydroxyl group of alcohol 11 could
be transformed to methyl, giving 13a-methyl antofine analogue
12 via a methanesulfonylation and superhydride reduction
consequence. Enantiopure 13a-methyl-14-substituted antofine
analogues were prepared as shown in Figure 5.32 From the
treatment of the readily available phenanthryl alcohol 13 with
bromine in DCM at room temperature, dibromide compound
14 was obtained in 85% yield. The other part 16 needed for the
following alkylation was prepared from optical oxazolidinone
ent-8 through modified Seebach’s self-regeneration of stereo-
centers (SRS) strategy to introduce a methyl group, which was
followed by acid-promoted methylation and deprotection
sequences. The following coupling of compound 17 and
dibromide 14 was not so seemingly easy, because compound 16
was hygroscopic and dibromide 14 was liable to moisture. After
extensive screening of reaction conditions, including bases,
solvents, and reaction temperatures, the desired product 17
could be obtained in good yield when K2CO3 (5 equiv) was
employed as a base in refluxed mixed solvents (1:1 DCM/
DMF). Then, the key intermediate 17 was transformed to
pentacyclic compound 18 through the Parham cycloacylation
reaction. To achieve stereoselective reduction of ketone 18, a
variety of hydride reagents were examined. It was found that
when superhydride (LiEt3BH) was used as the reducing agent,
the best selectivity (8:2 19/20) was obtained. Using less steric
LiAlH4 as a hydride source, the opposite selectivity was
observed.
Table 1. In Vitro and in Vivo Anti-TMV Activity of
Synthesized Compounds
To judge the antiviral potency of the synthesized D-ring
opened phenanthrolizidine analogues 3, 4, 5a−5j, 6, and 7 and
chiral 13a- and/or 14-substituted phenanthroindolizidines
(10−12, 18−20, ent-10−ent-12, and ent-18−ent-20), ribavir-
in and Ningnanmycin, two successful commercial plant
virucides, were selected as the controls. To further investigate
the influence of the substitutions at the 13a position and in-
depth SARs under the chiral scenario, enantiopure 13a-H
phenanthroindolizidines (S)-antofine, (R)-antofine, and race-
mic 13-methyl antofine were also used for comparison, which
were prepared by our previously reported methods.26 The in
vitro and in vivo (under three models: protection, inactivation,
and curative effect) activities against TMV of phenanthroindo-
lizidine alkaloid analogues, ribavirin, and Ningnanmycin were
shown in Table 1.
The in vitro anti-TMV bioassay showed that most of the
synthesized compounds except for compounds 6 and 10
exhibited higher or comparable anti-TMV activity compared to
ribavirin at both high (500 μg mL−1) and low (100 μg mL−1)
concentrations. Importantly, almost half of the compounds
(e.g., 4, 5j, 11, 19, 20, ent-10, ent-11, ent-12, and ent-20)
exhibited a similar inhibitory effect against TMV with
Ningnanmycin, perhaps the most effective commercial plant
virucide. Because most of the synthesized antofine analogues
displayed distinguished in vitro antiviral activity against TMV,
further investigations of their antiviral activity in vivo were
deserved. Then, in vivo activities were assayed in three models
(protection, deactivation, and curative) (Table 1).
Further in vivo anti-TMV activities showed that these new
synthesized phenanthroindolizidine alkaloid analogues dis-
played a good to excellent inhibitory effect, which, in general,
exhibited the same trend to the in vitro activities. Especially,
compound ent-11 showed a much more potent inhibitory
effect against TMV than the most effective commercial virucide
Ningnanmycin. Interestingly, although compounds 11, 19, 20,
ent-10, ent-12, and ent-20 exhibited inferior activity at 500 μg
mL−1 compared to Ningnanmycin, they showed higher or
comparable inhibitory effect at 100 μg mL−1. In other words,
these compounds could still keep very potent antiviral activities
even at a low concentration. Although most of the other
compounds showed less potent inhibitory effects than
2043
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045

Journal of Agricultural and Food Chemistry
Article
Ningnanmycin, they displayed better or at least comparable
activities than the other popular virucide ribavirin that was used
as a control.
Detailed synthetic procedure (PDF)
AUTHOR INFORMATION
■
SARs demonstrated that, although, as the controls, the
naturally occurring 13a-H phenanthrylindolizidine alkaloid (R)-
antofine showed comparable antiviral activities at both high and
low concentrations with its unnatural antipode (S)-antofine, all
of the synthesized 13a-substituted phenanthrylindolizidine
analogues ent-10, ent-11, and ent-12 exhibited more potent
antiviral effect than their corresponding configurationally
enantiomers 10, 11, and 12, respectively. These interesting
results obviously demonstrated that spatial configuration of the
molecules was very crucial for keeping high bioactivities and
introduction of suitable functional groups at the 13a position
could accelerate this trend. Both compounds 18 and ent-18
possessing a ketone group on the six-membered D ring, which
makes the D ring of the molecules more planar and, thus,
increases the molecular rigidity, were less effective than
compounds 12 and ent-12. After reduction of the ketones 18
and ent-18, the high rigidity of the molecule was released to
some extent and the resulting 13a-methyl-14-hydroxyl phenan-
throindolizidine analogues (19, 20, ent-19, and ent-20)
displayed higher inhibitory effect against TMV. However,
more interestingly, most of the D-ring opened analogues, which
became more flexible spatially, could also show comparable or
even higher (5a and 5j) anti-TMV activities than 13a-methyl
antofine. These results indicated that 3D conformation of
phenanthroindolizidine analogues was also an important factor
that can influence bioactivities. Another interesting observation
was that the hydroxyl group at either the 13a or 14 position was
helpful for keeping high biological activities; for example,
compounds 11, 19, 20, ent-11, ent-19, and ent-20 showed
more effective or at least a comparable inhibitory effect
compared to the 13a-methyl analogues 12 and ent-12, which
may suggest that a hydrogen-donating hydroxyl group at this
domain was favorable.
In summary, to explore the influence of spatial configuration
and 3D conformation of phenanthroindolizidine alkaloid
analogues on the activities, a series of D-ring opened
phenanthroindolizidine alkaloid analogues and chiral 13a-
and/or 14-substituted analogues were synthesized and
evaluated for their antiviral activities against TMV. The
bioassay results indicated that most of the chiral phenan-
throindolizidine analogues showed good to excellent in vivo
anti-TMV activity, among which compound 11 displayed a
more potent inhibitory effect than Ningnanmycin, thus
emerging as a potential inhibitor of the plant virus. An in-
depth SAR investigation showed that substitution patterns at
13a and/or 14 positions have a great effect on the anti-TMV
activity, demonstrating that spatial configuration is of
importance for keeping high activity. Molecular planarity and
rigidity of the D ring also greatly influence activity, showing
that 3D conformation is also very crucial for improving
bioactivity. Then, it is noteworthy that a hydrogen donor at the
13a or 14 position may be another essential factor for antiviral
investigation. Further investigations on structural optimization
and mode of action are currently underway in our laboratories.
Corresponding Author
*Telephone/Fax: +86-022-23503952. E-mail: wangqm@
nankai.edu.cn.
Funding
The authors are grateful to the National Natural Science
Foundation of China (21132003, 21421062, and 21372131)
and the Specialized Research Fund for the Doctoral Program of
Higher Education (20130031110017) for generous financial
support for their research programs.
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) Scholthof, K.-B. G. TOBACCO MOSAIC VIRUS: A Model
System for Plant Biology. Annu. Rev. Phytopathol. 2004, 42, 13−34.
(2) Hari, V.; Das, P. Ultra microscopic detection of plant viruses and
their gene products. In Plant Disease Virus Control; Hadidi, A.,
Khetarpal, R. K., Koganezawa, H., Eds.; APS Press: St. Paul, MN,
1998; pp 417−427.
(3) Bos, L. 100 years of virology: from vitalism via molecular biology
to genetic engineering. Trends Microbiol. 2000, 8, 82−87.
(4) Ouyang, G. P.; Cai, X. J.; Chen, Z.; Song, B. A.; Bhadury, P. S.;
Yang, S.; Jin, L. H.; Xue, W.; Hu, D. Y.; Zeng, S. Synthesis and antiviral
activities of pyrazole derivatives containing an oxime moiety. J. Agric.
Food Chem. 2008, 56, 10160−10167.
(5) Reichman, M.; Devash, Y.; Suhadolnik, R. J.; Sela, I. Human
leukocyte interferon and the antiviral factor (AVF) from virus-infected
plants stimulate plant tissues to produce nucleotides with antiviral
activity. Virology 1983, 128, 240−244.
(6) Chen, M. H.; Chen, Z.; Song, B. A.; Bhadury, P. S.; Yang, S.; Cai,
X. J.; Hu, D. Y.; Xue, W.; Zeng, S. Synthesis and antiviral activities of
chiral thiourea derivatives containing an α-aminophosphonate moiety.
J. Agric. Food Chem. 2009, 57, 1383−1388.
(7) Li, Y. M.; Zhang, Z. K.; Jia, Y. T.; Shen, Y. M.; He, H. P.; Fang, R.
X.; Chen, X. Y.; Hao, X. J. 3-Acetonyl-3-hydroxyoxindole: A new
inducer of systemic acquired resistance in plants. Plant Biotechnol. J.
2008, 6, 301−308.
(8) Sidwell, R. W.; Huffman, J. H.; Khare, G. P.; Allen, L. B.;
Witkowski, J. T.; Robins, R. K. Broad-spectrum antiviral activity of
virazole: 1-β-D-Ribofuranosyl-1,2,4-triazole-3-carboxamide. Science
1972, 177, 705−706.
(9) Bachrach, U. Antiviral activity of oxidized polyamines. Amino
Acids 2007, 33, 267−272.
(10) De Meester, C. Genotoxic potential of β-carbolines: A review.
Mutat. Res., Rev. Genet. Toxicol. 1995, 339, 139−153.
(11) Qian, X. H.; Lee, P. W.; Cao, S. China: forward to the green
pesticides via a basic research program. J. Agric. Food Chem. 2010, 58,
2613−2623.
(12) Seiber, J. N. Sustainability and agricultural and food chemistry. J.
Agric. Food Chem. 2011, 59, 1−2.
(13) Gellert, E. The indolizidine alkaloids. J. Nat. Prod. 1982, 45, 50−
73.
(14) Gellert, E.; Rudzats, R. The antileukemia activity of tylocrebrine.
J. Med. Chem. 1964, 7, 361−362.
(15) Damu, A. G.; Kuo, P. C.; Shi, L. S.; Li, C. Y.; Kuoh, C. S.; Wu, P.
L.; Wu, T. S. Phenanthroindolizidine alkaloids from the stems of Ficus
septica. J. Nat. Prod. 2005, 68, 1071−1075.
(16) Yang, C. W.; Chen, W. L.; Wu, P. L.; Tseng, H. Y.; Lee, S. J.
Antiinflammatory mechanisms of phenanthroindolizidine alkaloids.
Mol. Pharmacol. 2006, 69, 749−758.
(17) Baumgartner, B.; Erdelmeier, C. A. J.; Wright, A. D.; Rali, T.;
Sticher, O. An antimicrobial alkaloid from Ficus septica. Phytochemistry
1990, 29, 3327−3330.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jafc.5b06112.
2044
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045

Journal of Agricultural and Food Chemistry
Article
(18) Choi, J. Y.; Gao, W.; Odegard, J.; Shiah, H. S.; Kashgarian, M.;
McNiff, J. M.; Baker, D. C.; Cheng, Y. C.; Craft, J. Abrogation of skin
disease in LUPUS-prone MRL/FASIpr mice by means of a novel
tylophorine analog. Arthritis Rheum. 2006, 54, 3277−3283.
(19) An, T. Y.; Huang, R. Q.; Yang, Z.; Zhang, D. K.; Li, G. R.; Yao,
Y. C.; Gao, J. Alkaloids from Cyanachum komarovii with inhibitory
activity against the tobacco mosaic virus. Phytochemistry 2001, 58,
1267−1269.
(20) Xi, Z.; Zhang, R. Y.; Yu, Z. H.; Ouyang, D. The interaction
between tylophorine B and TMV RNA. Bioorg. Med. Chem. Lett. 2006,
16, 4300−4304.
(21) Pettit, G. R.; Goswami, A.; Cragg, G. M.; Schmidt, J. M.; Zou, J.-
C. Antineoplastic Agents, 103. The Isolation and Structure of
Hypoestestatins 1 and 2 From the East African Hypoestes verticillaris.
̈
J. Nat. Prod. 1984, 47, 913−919.
(22) Ali, M.; Bhutani, K. K. Minor Alkaloids of Tylophora hirsuta.
Phytochemistry 1987, 26, 2089−2092.
(23) Ali, M.; Bhutani, K. K. Alkaloids From Tylophora indica.
Phytochemistry 1989, 28, 3513−3517.
(24) Wang, K.; Su, B.; Wang, Z.; Wu, M.; Li, Z.; Hu, Y.; Fan, Z.; Mi,
N.; Wang, Q. Synthesis and Antiviral Activities of Phenanthroindo-
lizidine Alkaloids and Their Derivatives. J. Agric. Food Chem. 2010, 58,
2703−2709.
(25) Wang, Z.; Wei, P.; Xizhi, X.; Liu, Y.; Wang, L.; Wang, Q. Design,
Synthesis, and Antiviral Activity Evaluation of Phenanthrene-Based
Antofine Derivatives. J. Agric. Food Chem. 2012, 60, 8544−8551.
(26) Wang, Z.; Wei, P.; Wang, L.; Wang, Q. Design, Synthesis, and
Anti-tobacco Mosaic Virus (TMV) Activity of Phenanthroindolizidines
and Their Analogues. J. Agric. Food Chem. 2012, 60, 10212−10219.
(27) Wang, Z.; Wang, L.; Ma, S.; Liu, Y.; Wang, L.; Wang, Q. Design,
Synthesis, Antiviral Activity, and SARs of 14-Aminophenanthroindo-
lizidines. J. Agric. Food Chem. 2012, 60, 5825−5831.
(28) Wu, M.; Han, G.; Wang, Z.; Liu, Y.; Wang, Q. Synthesis and
Antiviral Activities of Antofine Analogues with Different C-6
Substituent Groups. J. Agric. Food Chem. 2013, 61, 1030−1035.
(29) Su, B.; Chen, F.; Wang, L.; Wang, Q. Design, Synthesis,
Antiviral Activity, and Structure−Activity Relationships (SARs) of
Two Types of Structurally Novel Phenanthroindo/quinolizidine
Analogues. J. Agric. Food Chem. 2014, 62, 1233−1239.
(30) Su, B.; Cai, C.; Deng, M.; Liang, D.; Wang, L.; Wang, Q. Design,
synthesis, antiviral activity, and SARs of 13a-substituted phenan-
throindolizidine alkaloid derivatives. Bioorg. Med. Chem. Lett. 2014, 24,
2881−2884.
(31) Su, B.; Cai, C.; Wang, Q. Enantioselective Approach to 13a-
Methylphenanthroindolizidine Alkaloids. J. Org. Chem. 2012, 77,
7981−7987.
(32) Su, B.; Deng, M.; Wang, Q. The First Enantioselective Approach
to 13a-Methyl-14-hydroxyphenanthroindolizidine AlkaloidsSyn-
thetic Studies towards Hypoestestatin 2. Eur. J. Org. Chem. 2013,
2013, 1979−1985.
(33) Su, B.; Cai, C. L.; Deng, M.; Liang, D. M.; Wang, L. Z.; Wang,
Q. M. Design, synthesis, antiviral activity, and SARs of 13a-substituted
phenanthroindolizidine alkaloid derivatives. Bioorg. Med. Chem. Lett.
2014, 24, 2881−2884.
2045
DOI: 10.1021/acs.jafc.5b06112
J. Agric. Food Chem. 2016, 64, 2039−2045