New rubrolide analogues as inhibitors of photosynthesis light reactions

Article abstract of DOI:10.1016/j.jphotobiol.2015.02.016

Natural products called rubrolides have been investigated as a model for the development of new herbicides that act on the photosynthesis apparatus. This study comprises a comprehensive analysis of the photosynthesis inhibitory ability of 27 new structura

Full text of DOI:10.1016/j.jphotobiol.2015.02.016

Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology B: Biology
journal homepage: www.elsevier.com/locate/jphotobiol
New rubrolide analogues as inhibitors of photosynthesis light reactions
a
a,b,
⇑
c
a
Jodieh O.S. Varejão , Luiz C.A. Barbosa
, Gabriela Álvarez Ramos , Eduardo V.V. Varejão ,
c
c
Beatriz King-Díaz , Blas Lotina-Hennsen
a
Department of Chemistry, Universidade Federal de Viçosa, Av. P.H. Rolfs, s/n, 36570-000 Viçosa, MG, Brazil
Department of Chemistry, Universidade Federal de Minas Gerais, Av. Pres. Antônio Carlos, 6627, Campus Pampulha, CEP 31270-901 Belo Horizonte, MG, Brazil
Department of Biochemistry, School of Chemistry, National Autonomous University of México, México, D.F. C.P. 04510, Mexico
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Natural products called rubrolides have been investigated as a model for the development of new herbi-
cides that act on the photosynthesis apparatus. This study comprises a comprehensive analysis of the
photosynthesis inhibitory ability of 27 new structurally diverse rubrolide analogues. In general, the
results revealed that the compounds exhibited efficient inhibition of the photosynthetic process, but in
some cases low water solubility may be a limiting factor. To elucidate their mode of action, the effects
of the compounds on PSII and PSI, as well as their partial reaction on chloroplasts and the chlorophyll
a fluorescence transients were measured. Our results showed that some of the most active rubrolide ana-
Received 4 December 2014
Received in revised form 9 February 2015
Accepted 19 February 2015
Available online 26 February 2015
B 1
logues act as a Hill reaction inhibitors at the Q level by interacting with the D protein at the reducing
side of PSII. All of the active analogues follow Tice’s rule of 5, which indicates that these compounds pre-
sent physicochemical properties suitable for herbicides.
Ó 2015 Elsevier B.V. All rights reserved.
1
. Introduction
In recent decades, increasing agricultural productivity has been
demand, academic and industrial studies have focused on the
search for natural products as models for the development of novel
herbicides [7–9].
a necessary and challenging task. The production of food for a
growing world population coupled to the allocation of cropland
to biofuel production without expanding the agricultural frontier
into natural ecosystems requires increasing agricultural yields
per hectare. Therefore, the reduction of any crop losses to agricul-
tural pests becomes a great concern.
Because weeds are one of the major pests causing agricultural
losses worldwide, the maintenance of high agricultural yields is
highly dependent on the chemical control of such invasive plants
In line with this tendency, our research group has synthesized
and investigated the phytotoxic activity of a range of analogues
to naturally occurring c-alkylidene butenolides [10–16]. In a recent
work, we found that synthetic analogues of rubrolides, namely a
class of natural butenolides, were almost as effective as the com-
mercial herbicide DCMU in inhibiting the photosynthetic electron
transport chain [10]. In the same study [10], a quantitative struc-
ture–activity relationship (QSAR) analysis showed that the photo-
synthesis-inhibitory activity of these compounds is strongly
related to their ability to accept electrons and with their polarity.
Through cyclic voltammetry and computational methods, we
found that the ability of a series of rubrolide analogues to interfere
with the Hill reaction is closely related to their redox potential: an
inhibitor with a higher first-reduction potential is associated with
increased effectiveness [11].
In this manuscript, we report a more comprehensive study that
was performed to obtain in-depth insights on how different substi-
tution patterns on the rubrolide analogues may influence their
photosynthesis-inhibitory properties. The range of compounds
tested includes both previously synthesized molecules [11] and
novel analogues, the synthesis and characterization of which are
also reported in this manuscript.
[
1,2]. However, the continuous application of herbicides has result-
ed in selective pressure, leading to the replacement of sensitive
biotypes by herbicide-resistant biotypes. The shift in the weed
spectrum in diverse croplands has made weed management an
even more difficult task [3–5]. Compounding this situation, no her-
bicide with a major new mode of action has been introduced into
the market in the last two decades [6]. In view of this scenario, it is
imperative to discover new herbicide molecules that are capable of
acting through novel mechanisms of action and additionally pre-
sent benign environmental and toxicological profiles. To meet this
⇑
Corresponding author at: Department of Chemistry, Universidade Federal de
Minas Gerais, Av. Pres. Antônio Carlos, 6627, Campus Pampulha, CEP 31270-901
Belo Horizonte, MG, Brazil. Tel.: +55 31 34093396.
With the aim of obtaining a better understanding of their mode
of action, the effects of the rubrolide analogues on various
E-mail addresses: lcab@ufmg.br, lcab@outlook.com (L.C.A. Barbosa).
http://dx.doi.org/10.1016/j.jphotobiol.2015.02.016
1
011-1344/Ó 2015 Elsevier B.V. All rights reserved.

1
2
J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
photosynthetic processes, including the electron transport rate
basal, phosphorylating and uncoupled), partial reactions of PSI
(47), 211 (48), 136 (54), 133 (22), 132 (22), 131 (35), 108 (100),
107 (74), 57 (26).
(
and PSII and the chlorophyll a fluorescence transients, were mea-
sured. The influence of the structural diversity of the rubrolide ana-
logues on their ability to interfere with these different
photosynthetic processes are discussed in the manuscript.
(Z)-3,4-Dibromo-5-(4-(trifluoromethyl)benzylidene)furan-2(5H)-
one, 6. The compound was obtained as a yellow solid at 31% yield.
ꢀ1
Melting point (mp) 137.0–137.8 °C. FT-IR (KBr)
1792, 1614, 1551, 1417, 1323, 1186, 1181, 1107, 1163, 980, 875.
mmax cm 3073,
ꢀ
1
H NMR d
.73 (d, 2H, J = 8.4 Hz). C NMR d
H
(300 MHz, CDCl
3
) 6.46 (s, 1H), 7.25 (d, 2H, J = 8.4 Hz),
(75 MHz, CDCl ) 112.0, 114.5,
1
3
7
H
3
2
. Materials and methods
125.9 (q, J = 3.8 Hz), 128.5 (q, J = 267.8 Hz), 130.9, 131.0 (q,
+
J = 32.3 Hz), 135.7, 137.4, 146.9, 162.8. MS, m/z (%) 400 (M +4,
+
+
2.1. General experimental
46), 398 (M +2, 100), 396 (M , C12
60), 186 (70), 158 (82), 133 (27), 131 (32), 89 (23), 86 (22), 84
(31), 63 (26), 51 (25).
5 2 3 2
H Br F O , 48), 263 (53), 261
(
All of the reagents and solvents were prepared following proce-
dures already reported in the literature [17] or were purchased
from commercially available suppliers and used without any fur-
ther purification. The reported melting points are uncorrected
and were obtained using a MQAPF-301 melting point apparatus
(Z)-3,4-Dibromo-5-(4-ethylbenzylidene)furan-2(5H)-one, 7. The
compound was obtained as a yellow solid at 47% yield. Melting
ꢀ
1
point (mp) 104.7–105.8 °C. FT-IR (KBr)
2929, 2871, 1782, 1601, 1205, 846, 739, 621. H NMR d
(300 MHz, CDCl
6.43 (s, 1H), 7.25 (d, 2H, J = 8.4 Hz), 7.80 (d, 2H, J = 8.4 Hz).
NMR d (75 MHz, CDCl
131.1, 137.4, 145.0, 147.1, 163.5. MS, m/z (%) 360 (M +4, 19), 358
m
ꢀ
max cm
3041, 2964,
1
H
(
Microquimica, Brazil). Analytical thin layer chromatography ana-
3
) 1.26 (t, 2H, J = 7.5 Hz), 2.68 (q, 3H, J = 7.5 Hz),
13
lysis was conducted on aluminum-packed pre-coated silica gel
plates. Column chromatography was performed over silica gel
C
H
3
) 15.2, 28.9, 112.4, 114.5, 128.6, 129.3,
+
2
30–400 mesh. All of the compounds were fully characterized by
1
13
+
+
IR, EI-MS, H NMR, C NMR, COSY, HETCOR and NOEDIFF NMR
spectroscopy. The infrared spectra were recorded on a Perkin–
Elmer Paragon 1000 FTIR spectrophotometer using potassium bro-
mide (1% w/w) disks or thin liquid film on NaCl plates. The mass
spectra were recorded on a Shimadzu GCMS-QP5050A instrument
by direct insertion using the EI mode (70 eV). The H NMR and
NMR spectra were recorded on a Varian Mercury 300 spectrometer
at 300 and 75 MHz, respectively, using CDCl or (CD CO as sol-
2 2
(M +2, 39), 356 (M , C13H10Br O , 19), 345 (20), 343 (35), 341
(23), 131 (100), 117 (21), 115 (29), 77 (41), 63 (26), 51 (24), 39
(30).
(Z)-3,4-Dibromo-5-(3-methoxybenzylidene)furan-2(5H)-one, 8.
The compound was obtained as a yellow solid at 67% yield. Melt-
1
13
ꢀ1
C
ing point (mp) 151.0–151.6 °C. FT-IR (KBr)
m
ꢀ
max cm 3063, 3044,
3015, 2967, 2941, 2839, 1765, 1649, 1596, 1549, 1307, 1237,
1
3
3
)
2
1191, 1040, 979, 880, 789, 684, 637. H NMR d
H 3
(300 MHz, CDCl )
vents and TMS as an internal reference, unless otherwise stated.
Compounds 2–29 were obtained by employing a synthetic proce-
dure previously described [10,11]. The spectroscopic assignments
for compounds 2 and 9–29 agreed with those reported in the lit-
erature [11,18]. The structures of compounds 3–8 are supported
by the spectroscopic data reported below.
3.85 (s, 3H), 6.41 (s, 1H), 6.96 (ddd, 3H, J = 7.8 Hz, J = 2.7 Hz,
1
3
J = 1.5 Hz), 6.97–7.39 (m, 3H).
H 3
C NMR d (75 MHz, CDCl )
55.4, 113.2, 114.1, 115.6, 116.4, 123.8, 130.1, 133.0, 137.5,
+
+
145.7, 159.8, 163.2. MS, m/z (%) 363 (M +4, 47), 360 (M +2,
+
8 2 2
100), 358 (M , C12H Br O , 50), 281 (37), 279 (45), 225 (53),
223 (56), 167 (24), 149 (80), 148 (57), 91 (46), 86 (31), 84 (51),
77 (38), 71 (38), 63 (23), 57 (69), 55 (39), 51 (90), 50 (30).
(
Z)-3-Bromo-5-(3-methoxybenzylidene)-4-phenylfuran-2(5H)-
one, 3. The compound was a yellow solid obtained at 70% yield.
ꢀ
1
Melting point (mp) 106.5–107.4 °C. FT-IR (KBr)
2
m
ꢀ
max cm 3013,
960, 2933, 2833, 1773, 1571, 1478, 1447, 1290, 1254, 1161,
049, 981, 874, 770, 745, 686. 1H NMR d
(300 MHz, CDCl ) 3.84
2.2. Chloroplasts isolation and chlorophyll determination
1
H
3
Intact chloroplasts were isolated from spinach leaves (Spinacia
oleracea L.) obtained from a local market as previously described
[19,20]. The chloroplasts were isolated with a medium that
(
s, 3H), 6.11 (s, 1H), 6.90–6.94 (m, 1H), 7.26–7.36 (m, 4H), 7.51–
(75 MHz, CDCl ) 55.7, 109.1, 114.9,
H 3
15.8, 116.3, 124.0, 129.3, 129.4, 130.2, 130.9, 134.1, 148.0,
54.2, 160.1, 165.3. MS, m/z (%) 358(M +2, 52), 356 (M , C18
, 52), 277 (31), 249 (31), 221 (100), 178 (25), 148 (36), 129
47), 125 (20), 91 (37), 89 (21), 77 (25), 51 (50).
Z)-3-Bromo-5-(4-fluorobenzylidene)-4-phenylfuran-2(5H)-one,
. The compound was obtained as a yellow solid at 44% yield. Melt-
1
3
7
1
1
.59 (m, 4H). C NMR d
2
contained 400 mM sucrose, 5 mM MgCl , and 10 mM KCl and
+
+
+
H
13
-
was buffered with 3 mM K -tricine at pH 8.0. The chloroplasts
were re-suspended in the same medium and stored as a concen-
trated suspension in the dark at 4.0 °C while using. The chlorophyll
concentration was measured spectrophotometrically as previously
reported [21].
BrO
2
(
(
4
ꢀ1
ing point (mp) 138.8–139.3 °C. FT-IR (KBr) mmax cm 3061, 1771,
ꢀ
1
600, 1508, 1239, 1163, 955, 830, 748, 697. 1H NMR
d
H
2.3. Measurement of non-cyclic electron transport rate
(
(
300 MHz, CDCl
m, 5H), 7.78 (dd, 2H, J = 5.4 Hz, J = 8.7 Hz). C NMR d
108.5, 113.3, 116.1 (d, J = 21.8 Hz), 128.8, 128.8 (d,
J = 2.3 Hz), 128.9, 130.0, 130.5, 132.8 (d, J = 8.3 Hz), 147.1, 153.8,
3
) 6.10 (s, 1H), 7.08 (t, 2H, J = 8.7 Hz), 7.50–7.59
13
H
(75 MHz,
The light-induced non-cyclic electron transport activity from
water to methylviologen (MV) was determined with a YSI (Yellow
Springs Instrument) oxygen monitor (model 5300) using a Clark-
type electrode as previously published [19,20]. The chloroplasts
3
CDCl )
+
+
1
63.2 (J = 255 Hz), 164.9. MS, m/z (%) 346 (M +2, 35), 344 (M , C17
10BrFO , 36), 209 (100), 136 (39), 129 (40), 108 (66), 107 (31), 75
23).
Z)-3,4-Dibromo-5-(4-fluorobenzylidene)furan-2(5H)-one, 5. The
compound was obtained as a yellow solid at 14% yield. Melting
-
H
(
2
(equivalent of 20
3 mL of the basal electron transport reaction medium, which was
composed of 100 mM sorbitol, 5 mM MgCl , 10 mM KCl, 0.5 mM
M MV, and the basal elec-
lg/mL of chlorophyll) were freshly lysed in
(
2
+
KCN, 30 mM K -tricine at pH 8.0 and 50
l
ꢀ1
point (mp) 178.5–180.0 °C. FT-IR (KBr)
603, 1547, 1511, 1247, 1166, 992, 966, 826. H NMR d
300 MHz, CDCl
m
ꢀ
max cm
3059, 1767,
tron flow was measured by polarography. The phosphorylating
non-cyclic electron transport from water to MV was measured
with basal non-cyclic electron transport medium supplemented
1
1
(
2
1
H
3
) 6.41 (s, 1H), 7.11 (t, 2H, J = 8.7 Hz), 7.81 (dd,
13
H, J = 5.4 Hz, J = 8.7 Hz). C NMR d
H
3
(75 MHz, CDCl ) 112.9,
2 4
with 1 mM ADP and 3 mM KH PO . The uncoupled electron trans-
13.0, 116.3 (d, J = 21.8 Hz), 128.1 (d, J = 3.8 Hz), 133.1 (d,
port from water to MV was tested using the same protocol as that
used for the measurement of the basal non-cyclic electron trans-
port with the exception that 6 mM NH Cl was added as an uncou-
4
J = 8.3 Hz), 137.4, 145.2, 163.3, 163.5 (J = 252 Hz). MS, m/z (%)
+
+
+
3
50 (M +4, 36), 348 (M +2, 74), 346 (M , C11
5
H Br
2 2
FO , 36), 213

J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
13
pler to the medium. All of the reaction mixtures were illuminated
for 2 min with a projector lamp (Gaf 2669) and passed through a
concentration. The appearance of a K-band near 0.3 ms is an indi-
cation of damage in the water splitting system on thylakoids [27].
5
4
cm filter of a 1% CuSO solution to result in actinic light
2
(
0.2 mW/cm ).
3
. Results and discussion
2
.4. Determination of uncoupled photosystem II (PSII) and uncoupled
3.1. Synthesis
photosystem I (PSI) electron flow
All of the compounds were prepared using the same synthetic
The uncoupled PSII from water to 2,5-dichloro-1,4-benzo-
quinone (DCBQ) was monitored polarographically [22]. DCBQ
approach previously reported [10,11,28]. Briefly, brominated lac-
tone 1 was subjected to aldol condensation using different aromat-
ic aldehydes, tert-butyldimethylsilyl trifluoromethane sulfonate
TBDMSOTf) and N,N-diisopropylethylamine (DIPEA) (Fig. 1), to
afford the adduct, which was not isolated. Then, the addition of
,8-diazabicyclo[5.4.0] undec-7-ene (DBU) to the reaction media
led to a b-elimination and the formation of compounds 5–8 at
yields ranging from 14% to 67%. For the synthesis of 3 and 4, lac-
tone 2 was first synthesized (66% yield) through a well-established
methodology [18] utilizing a Suzuki–Miyaura cross coupling
between brominated lactone 1 and phenylboronic acid. Lactone 2
was then subjected to aldol condensation with p-fluorobenzalde-
hyde and m-methoxybenzaldehyde. A subsequent b-elimination
led to compounds 3 and 4 at yields of 44% and 70%, respectively.
The synthesis of compounds 9–29 has been previously reported
accepts electrons at the D
tron transport from water to Q
PSII activity was the same as that used for basal electron transport
with the exception that MV was omitted and 1 M 2,5-dibromo-6-
isopropyl-3-methyl-1,4-benzoquinone (DBMIB), 100 M DCBQ and
mM NH Cl were added. The partial reaction of PSII electron trans-
1
protein, and thus, we measured the elec-
B
. The reaction medium for assaying
(
l
1
l
6
4
port from water to sodium silicomolybdate (SiMo) was determined
using the same protocol as that used for PSII with the same basal
electron transport medium without MV and supplemented with
5
0
lM SiMo and 10
lM 3-(3,4-dichlorophenyl)-1,1-dimethylurea
(
DCMU) [23]. These electron flow activities were monitored with
a YSI oxygen monitor (model5300) using a Clark-type electrode.
The uncoupled PSI electron transport from 2,6-dichloropheno-
lindophenol (DCPIP) reduced with ascorbate was determined using
[
11], and their substitution patterns are presented in Table 1.
the basal electron transport medium supplemented with 10
DCMU (to inhibit PSII at the Q level) [24], and 100 M DCPIP
was reduced with 300 M ascorbate and uncoupled with 6 mM
NH Cl [25]. The I50 value for each activity was extrapolated using
lM
B
l
l
3
.2. Effect of compounds 3–29 on electron transport from water to MV
in spinach chloroplasts
4
the graph of the percentage of activity as a function of the com-
pound concentration. I50 is the concentration resulting in 50% inhi-
bition of the activity.
To investigate their photosynthesis-inhibition properties, com-
pounds 3–29 were assayed for their effect on non-cyclic electron
transport from water to MV under basal, phosphorylation, and
uncoupled conditions.
2.5. Chlorophyll a fluorescence of PSII measurements in thylakoids
Compounds 8–10, 13–15, 18, 20, 22–24, 26 and 28 were inac-
tive, and were not further studied. The active compounds were list-
ed in Table 2, which show that all were capable of inhibiting the
basal, uncoupled and phosphorylating electron transfer, indicating
that these compounds act as Hill reaction inhibitors. The remark-
able variability in the activity of these compounds clearly indicates
that the presence and position of different substituents in aromatic
and lactone rings exert great influence on the ability of the result-
ing compounds to interfere with the photosynthetic electron flow.
As a general tendency, the presence of an electron-withdrawing
substituent, which improves the ability of the compound to accept
electrons, and hydroxyl groups, which makes the compound more
hydrosoluble, plays a pivotal role on the activities of these com-
pounds. This tendency becomes clearer when we compared those
compounds with closer structural similarities. For example, com-
pounds 11 and 12, 21 and 19, 17 and 14, and 25 and 23 shows
the same pattern of substitution on the benzylidene ring but differ
from each other by the presence of a fluorine atom at the para posi-
tion in the 3-phenyl ring. The comparison of these pairs of com-
pounds revealed that the rubrolide analogue bearing the
electron-withdrawing fluorine atom showed higher activity than
the corresponding analogue without this substituent. This influ-
ence is also observed for compounds 3 and 4: compound 3, which
bears a fluorine atom in the benzylidene ring, showed markedly
higher activity compared with 4, which has an electron-donating
methoxyl group on the same ring. A comparison of compounds 6,
5, 7 and 8 makes this tendency even more evident. These com-
pounds differ from each other only by the substituent on the ben-
zylidene ring. Compound 6 bears a strong electron-withdrawing
trifluoromethyl and showed the highest inhibitory activity. Com-
pound 5, which presents a fluorine atom, exhibited greater activity
in uncoupled electron flux compared with compound 7, which
The chlorophyll a fluorescence transients were measured with a
Handy-PEA (Plant Efficient Analyzer, from Hansatech, King’s Lynn,
Norfolk, UK) as previously described [20]. The maximum fluores-
cence yield from the sample was generated by illumination for
2
s with continuous light (650 nm peak wavelength, intensity
ꢀ
2 ꢀ1
equivalent to 2830
lmol photons m
s
and gain of 0.7) provided
by an array of three light-emitting diodes. The reaction medium
used was that employed in the basal non-cyclic electron transport
measurements with MV. To monitor the Chl a fluorescence tran-
sients, aliquots of dark-adapted thylakoids containing 60 lg of
Chl were incubated for 5 min in 3 mL of the basal electron trans-
port medium without MV and with different concentrations of
the tested compounds, and the control contained medium plus
the amount of DMSO employed for each compound at the different
concentrations tested. The solutions were then centrifuged at
1
269g for 5 min. The pellets containing the thylakoids were trans-
2
ferred to 1.5 cm of filter paper by gravity with a dot-blot appara-
tus (Bio-Rad, United States) to ensure homogeneous and
a
reproducible distribution of the thylakoids. The filter paper was
immediately placed in a clip, and the fluorescence was measured.
The OJIP transients were analyzed according to the JIP test, and
the measured parameters were the following: fluorescence intensi-
ty level (F
fully oxidized and fluorescence level when Q
reduced (F ). F = 0.05 ms, F = F (or F300us); F
= F (30 ms) [26].
The chlorophyll a fluorescence transients normalized using the
equation Vt = (Ft – F )/F – F ), and the difference between the
0
) when the plastoquinone electron acceptor pool (Q
is transiently fully
= F (2 ms);
A
) is
A
M
1
3
K
4
J
F
5
I
0
M
0
treated and control normalized transients are plotted as a function
of time, an increase in intensity between 2 and 4 ms is observed in
ꢀ
a
J-band close to 2 ms indicates an increase in the
Q
A

1
4
J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
Br
Br
Br
Br
PhB(OH)2
(1) ArCHO,TBDMSOTf
DIPEA, CH Cl
O
O
2
2
O
O
O
1
O
PdCl2(MeCN)2, AsPh3
Ag2O, THF, 65 ºC
(
2) DBU, CH Cl , reflux
2: 66%
2 2
R
(
1) ArCHO,TBDMSOTf
3
: R = p-F (44%)
DIPEA, CH Cl
2
2
4: R = m-OMe (70%)
(
2) DBU, CH Cl , reflux
2 2
Br
Br
O
O
R'
5
: R = p-F (14%)
6
7
8
: R = p-CF (31%)
3
: R = p-Et (47%)
: R = m-OMe (67%)
Fig. 1. Synthesis of compounds 3–8.
Table 1
Substituent groups of compounds 9–29.
observed for the new series of analogues investigated in this study,
corroborating the correlation between redox potential and ability
to act as a Hill reaction inhibitor [11].
The comparison of the pairs of compounds 17/15, 19/18, 21/20,
25/24, 27/26 and 29/28 revealed that the compounds presenting
methoxyl groups were less active compared with those bearing
hydroxyl groups. Thus, we can hypothesize that higher water solu-
bility, which is conferred by the presence of hydroxyl groups, may
influence the activity of these compounds.
R1
R2
Br
O
O
R6
R3
This trend was also found for nostoclide analogues, another
class of
c-alkylidenebutenolides with photosynthesis-inhibition
R5
R4
properties: the efficacy of nostoclides as Hill reaction inhibitors
was correlated with higher water solubility. As argued by the
authors, higher water solubility may improve the partition
between water and thylakoid membranes and consequently access
to the active site [14]. It is worth mentioning that all rubrolide ana-
logues presenting only methoxyl groups were inactive, whereas
almost all hydroxylated analogues interfered with the photosyn-
thetic electron flow (Table 2).
9
-29
Compound
R
1
R
2
R
3
R
4
R
5
R
6
9
H
F
H
F
H
H
F
F
F
H
H
F
OMe
OMe
OMe
OMe
OMe
OH
OMe
OMe
OH
OMe
OH
OMe
OH
OMe
OH
OMe
OH
OMe
OH
OMe
OH
H
H
H
H
OMe
OH
OMe
OH
OH
H
H
H
H
H
H
H
H
H
Et
Et
H
H
H
H
H
H
H
H
H
H
H
H
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
OH
OH
OMe
OH
OMe
OH
OH
OMe
OH
OMe
OH
H
H
H
H
H
H
H
H
H
Table 2
OMe
OH
OMe
OH
H
H
H
H
OMe
OH
OMe
OH
I
50 values (lM) of active compounds calculated from the basal, phosphorylating and
H
H
H
uncoupled electron transport from water to MV in spinach thylakoids.
F
H
Compound Basal I50
(lM) Phosphorylating I50
Uncoupled I50 (lM)
H
H
F
OMe
OH
OMe
OH
OMe
OH
OMe
OH
OMe
OH
OMe
OH
H
H
H
H
(lM)
3
4
5
6
7
1
2
6
7
9
1
4.5 ± 2.1
65 ± 8.6
35 ± 2.5
0.133 ± 0.04
22 ± 11
40 ± 3.5
4.0 ± 2.2
95 ± 5.7
63 ± 2.8
3.3 ± 1.7
43 ± 6.8
17 ± 1.8
0.536 ± 0.12
26 ± 4.3
79 ± 6.2
23 ± 3.4
1.8 ± 0.8
9.7 ± 3.2
99 ± 14.7
63 ± 1.7
14 ± 6.2
63 ± 2.6
69 ± 3.6
F
H
H
F
0.339 ± 0.1
21 ± 9.0
34 ± 5.7
58 ± 4.5
6.2 ± 5.8
4.3 ± 2.0
70 ± 5.4
56 ± 4.6
14 ± 6.8
78 ± 4.2
50 ± 4.5
1
1
1
1
1
2
F
33 ± 2.2
2.0 ± 1.2
2.5 ± 2.5
181 ± 10.5
91 ± 3.1
17 ± 7.1
48 ± 3.2
bears an ethyl substituent. Finally, compound 8, bearing an
electron-donating methoxyl group, was found to be inactive.
In a previous study [10], we found that rubrolide analogues
bearing electron-withdrawing substituents were more effective
in interfering with the Hill reaction. This finding was again
25
27
29
175 ± 9.6

J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
15
Table 3
translocation to the site of action, are encompassed by ‘‘Tice’s rule’’
for screening new herbicide candidates [31].
Tice’s parameters calculated for the active compounds.
Compound
milogP
MW
HBD
HBA
nRotb
To determine whether the photosynthesis-inhibitory rubrolide
analogues satisfy Tice’s rule, their physicochemical parameters
were calculated using the Molinspiration software package [32],
and these are summarized in Table 3. As observed, all of the com-
pounds follow Tice’s rule, strongly indicating that these com-
pounds present physicochemical properties suitable for
herbicides [31].
3
4
5
6
7
4.747
4.616
3.741
4.493
4.512
4.113
4.252
4.169
3.893
3.348
3.487
3.893
3.553
3.692
65
345.167
357.203
347.965
397.972
358.029
373.202
391.192
407.191
393.164
375.174
393.164
393.164
375.174
393.164
150–500
0
0
0
0
0
1
1
2
3
3
3
3
3
3
63
2
3
2
2
2
3
3
5
5
5
5
5
2
3
1
2
2
4
4
3
2
2
2
2
2
2
612
1
1
1
1
1
2
2
2
2
1
2
6
7
9
1
5
7
9
3.4. Localization of target sites of rubrolide analogues
To determine the target sites on the thylakoid electron trans-
5
5
2–12
port chain, the effect of rubrolide analogues 3, 6, 12, 16, 17 and
5 on PSII, PSI, and partial reactions of photosystems were tested
2
29
Tice
using artificial electron donors and acceptors, as well as appropri-
ate inhibitors [25]. As shown in Fig. 2A and B, compounds 3, 6, 12,
1
6, 17 and 25 completely inhibited the uncoupled PSII electron
flow from water to DCBQ, with I50 values of 14.5, 0.65, 3.7, 3.1,
.3 and 2.4 M, respectively.
To further identify the inhibitory site of the rubrolide analogues
on PSII, we measured their effects on the electron flow from water
to SiMo (because SiMo accepts electrons from pheophytin, the
electron flow measurements were performed from the water-split-
ting enzyme to pheophytin). The results indicate that the com-
pounds did not affect this partial reaction because the rate of the
treated thylakoids was similar to the control (data not shown);
therefore, all of the active rubrolide analogues assayed inhibited
the electron transport chain after pheophytin [23].
Compounds 19 and 29 inhibit the phosphorylating electron
transport rate to a higher degree than the basal and the uncoupled
electron flow (high I50 values from the basal electron flux were cal-
8
l
culated for these compounds: 181 lM and 175 lM, respectively).
These results indicate that 19 and 29 interact more at an energized
state (electron flow coupled to ATP synthesis) than in a non-ener-
gized state (basal and uncoupled electron flow) or may act by
inhibiting ATP synthesis itself [29].
The uncoupled PSI electron transport rate was measured by
polarography from reduced DCPIP to MV using increasing concen-
trations of the rubrolide analogues to determine whether they
affect this photosystem. The results showed that the rates for the
control and treated samples were similar. Therefore, the analogues
do not interact with the PSI electron transport chain (data not
shown).
3
.3. Predicted physicochemical parameters for active compounds
In a living organism, an active compound needs to cross several
barriers before reaching its target site. Thus, its biological activity
is greatly determined not only by its potency but also by its solu-
bility and ability to permeate cell walls and membranes, which
depends on the physicochemical properties of the compound
The polarographic measurements suggest that compounds 3, 6,
[
30,31].
‘Lipinski’s rule of 5’’ [30] is a set of empirically derived rules
1
2, 16, 17 and 25 inhibited electron transport at the acceptor side
‘
of PSII, since the rubrolide analogues have no-effect on electron
transport from water to pheophytin and on electron flow of PSI.
that delineate the physicochemical properties and molecular fea-
tures of orally bioavailable drugs. Molecules violating more than
one of these rules exhibited limited bioavailability [30]. This
approach was modified by Tice [31] for agrochemicals. The physi-
cochemical parameters molecular weight (MW), calculated lipo-
philicity (LogP), number of hydrogen-bond donors (HBD),
number of hydrogen-bond acceptors (HBA), and number of
rotatable bonds (nRotb), which is consistent with uptake and
3.5. Measurements of the Chl a fluorescence transients in thylakoids in
the presence of the synthetic rubrolide analogues
To provide additional evidence of the interaction site of the
rubrolide analogues at PSII, the fluorescence of chlorophyll a on
1
00
100
B
80
A
8
6
4
2
0
0
0
0
0
60
40
20
0
0
5
10
15
20
25
30
0
20
40
60
80
[
Compounds] µM
[Compounds] µM
Fig. 2. Panel A: Effect of 6 (h), 16 (d), 17 (j) and 25 (N). Panel B: Effect of 3 (4) and 12 (s). The effects were measured on the uncoupled PSII electron transport from H
2
O to
ꢀ
DCBQ. The control is equal to 100% of the activity and was 311
l
equiv e /mg Chl h.

1
6
J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
2
1
1
000
600
200
K
J
I
P
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
A
B
8
4
00
00
0.01
0.1
1
10
100
1000
0.01
0.1
1
10
100
1000
Time (ms)
Time (ms)
2
1
1
000
600
200
C
1.0
D
0
0
0
0
0
.8
.6
.4
.2
.0
8
00
4
00
0.01
0.1
1
10
100
1000
0.01
0.1
10
100
1000
1
Time (ms)
Time (ms)
1
1
1
400
200
000
E
8
6
4
2
00
00
00
00
0.01
0.1
1
10
100
1000
Time (ms)
Fig. 3. Chl a fluorescence transients of the control (j) and samples treated with the following: Panel A, 3 (N) (50
M), 12 (h) (100 M) and 17 (s) (30 M). The transients shown in Panels B and D were normalized from F to F
M DCMU (d) and 0.8 M Tris (N).
l
M), 16 (d) (30
l
M) and 25 (4) (20
l
M); Panel C, 6 ( )
r
(
3
l
l
l
1
M
. Panel E, Chl a fluorescence transients of the control (j)
and samples treated with 10
l
Table 4
Experimental average values ± standard deviation of chlorophyll a fluorescence obtained for the control and treated chloroplasts with compounds 3, 6, 12, 16, 17 and 25. The data
present the results of four replicates ± standard deviation.
Compound (lM)
F
0
F
M
F
1
= 0.050 (ms)
F
3
= F
K
F
4
= F
J
F
5
= F
I
F
M
/F
0
V M
F /F
Control
354 ± 24
332 ± 33
454 ± 33
598 ± 45
410 ± 15
430 ± 19
394 ± 46
1627 ± 80
1379 ± 37
1820 ± 34
1444 ± 47
1446 ± 15
1452 ± 22
1736 ± 51
389 ± 24
363 ± 45
500 ± 66
642 ± 47
456 ± 17
481 ± 26
436 ± 56
558 ± 38
520 ± 59
739 ± 60
859 ± 79
698 ± 29
738 ± 60
646 ± 108
838 ± 46
787 ± 12
1142 ± 64
982 ± 20
4.596
4.153
4.00
2.415
3.527
3.377
4.406
0.783 ± 0.009
0.759 ± 0.017
0.750 ± 0.009
0.585 ± 0.039
0.716 ± 0.006
0.704 ± 0.003
0.716 ± 0.006
3
6
1
1
1
2
(50)
(3)
2 (100)
6 (30)
7 (30)
5 (20)
1175 ± 70
1236 ± 86
1175 ± 30
1222 ± 67
1010 ± 103
1393 ± 90
1351 ± 98
1320 ± 18
1360 ± 69
1237 ± 146
Control
DCMU (10)
305 ± 15
413 ± 21
1401 ± 70
1355 ± 68
319 ± 16
450 ± 22
488 ± 24
471 ± 39
808 ± 40
1191 ± 60
986 ± 49
1334 ± 67
4.593
3.28
0.782 ± 0.006
0.695 ± 0.013
freshly lysed spinach chloroplasts was measured. The control
chloroplast contained the amount of DMSO used for each treat-
ment, and the treated chloroplasts contained the compounds at
the concentrations at which they were found to completely inhibit
PSII through polarographic measurements. The chlorophyll a
induction curves of the control thylakoids (Fig. 3, Panels A and B)
showed an OJIP sequence similar to that previously described for
plants, green algae, and cyanobacteria [33].
The chloroplasts treated with the rubrolide analogues and the
M M
control samples showed different F levels (Table 4). The F levels
of the chloroplasts treated with 16 (Fig. 3A), 17 and 12 (Fig. 2C)
were decreased compared with the control; these compounds

J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
17
showed the appearance of the highest J-bands. In contrast, the F
level was increased by almost all of the compounds tested with
the exception of 3. Furthermore, the initial (F ) and the maximum
) values of the fluorescence rise were used to determine the
quantum yield of primary photochemistry Po = F /F = kP/
kP + kN)], where kP and kN are rate constants of photochemistry
and other losses of excitation energy, respectively, and F is the
variable component of fluorescence obtained by subtraction of F
from the F value [34]. The modification of the F and F levels
0
[3] S.B. Powles, Evolved glyphosate-resistant weeds around the world: lessons to
be learnt, Pest Manag. Sci. 64 (2008) 360–365.
[
4] H.J. Beckie, Herbicide–resistant weed management: focus on glyphosate, Pest
Manag. Sci. 67 (2011) 1037–1048.
0
(
F
M
[5] S.O. Duke, S.B. Powles, Glyphosate–resistant crops and weeds: now and in the
future, AgBioForum. 12 (2009) 346–357.
[
u
V
M
[
6] S.O. Duke, Why has no new herbicide modes of action appeared in recent
years?, Pest Manag Sci. 68 (2012) 502–512.
(
V
[7] F.E. Dayan, C.L. Cantrell, S.O. Duke, Natural products in crop protection, Bioorg.
Med. Chem. 17 (2009) 4022–4034.
0
[
8] C.L. Cantrell, F.E. Dayan, S.O. Duke, Natural products as sources for new
pesticides, J. Nat. Prod. 75 (2012) 1231–1242.
M
M
0
in the presence of DCMU would imply a change in the de-excita-
tion rate constants (kP and kN).
The chlorophyll a fluorescence transients from Fig. 3A and C
1 M
were normalized between F and F (Fig. 3B and D). In Fig. 3B
[9] R.D. Clark, A perspective on the role of quantitative structure-activity and
structure-property relationships in herbicide discovery, Pest Manag. Sci. 68
(
2012) 513–518.
[
10] L.C.A. Barbosa, A.J. Demuner, E.E.L. Borges, J. Mann, Synthesis and evaluation of
the plant growth regulatory activity of 8-oxabicyclo[3.2.1] oct-6-en-3-one
derivatives, J. Braz. Chem. Soc. 8 (1997) 19–27.
[11] J.O.S. Varejão, L.C.A. Barbosa, C.R.A. Maltha, M.R. Lage, M. Lanznaster, J.W.M.
Carneiro, G. Forlani, Voltammetric and theoretical study of the redox
properties of rubrolides analogues, Electrochim. Acta 120 (2014) 334–343.
[12] L.C.A. Barbosa, J.O.S. Varejão, D. Petrollino, P.F. Pinheiro, A.J. Demuner, C.R.
Maltha, G. Forlani, Tailoring nostoclide structure to target the chloroplastic
electron transport chain, ARKIVOC iv (2012) 15–32.
and D it is possible to observe an increment at 2 ms, which corre-
ꢀ
sponds to the J-band which is related to the accumulation of Q
species, indicating that the electron transport beyond Q
at the Q level similarly to that achieved with DCMU (Fig. 3E) [35].
DCMU displaces the secondary quinone acceptor, Q , from its bind-
A
A
is blocked
B
B
ing site at the D
redox active and prevents the reoxidation of Q
1
protein of photosystem II (PSII) [35]. DCMU is not
[
[
[
13] R.R. Teixeira, P.F. Pinheiro, L.C.A. Barbosa, J.W.M. Carneiro, G. Forlani, QSAR
modeling of photosynthesis-inhibiting nostoclide derivatives, Pest Manag. Sci.
ꢀ
A
by forward elec-
66 (2010) 196–202.
tron transport. Since the fluorescence rise is mainly determined
14] R.R. Teixeira, L.C.A. Barbosa, G. Forlani, D. Piló-Veloso, J.W.M. Carneiro,
Synthesis of photosynthesis-inhibiting nostoclide analogues, J. Agric. Food
Chem. 56 (2008) 2321–2329.
ꢀ
by the redox state of Q
A
[36], blocking its re-oxidation, both the
polarographic technique and the fluorescence of chlorophyll a indi-
cate that the rubrolide analogues interact with and inhibit the pho-
15] L.C.A. Barbosa, M.E. Rocha, R.R. Teixeira, C.R.A. Maltha, G. Forlani, Synthesis of
3
-(4-bromobenzyl)-5-(arylmethylene)-5H-furan-2-ones and their activity as
inhibitors of the photosynthetic electron transport chain, J. Agric. Food Chem.
5 (2007) 8562–8569.
tosynthesis system by displacing Q
. Conclusions
In this study, we found that not only the presence of electron-
B
.
5
[
16] L.C.A. Barbosa, A.J. Demuner, E.S. Alvarenga, A. Oliveira, B. King-Diaz, B. Lotina-
Hennsen, Phytogrowth- and photosynthesis-inhibiting properties of
nostoclide analogues, Pest Manag. Sci. 62 (2006) 214–222.
17] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemicals, fifty ed.,
Butterworth-Heinemann, Boston, 2003.
4
[
[
with drawing groups but also the hydrosolubility of rubrolide ana-
logues exert influence on the ability of these compounds to act as
photosynthesis inhibitors. These findings corroborate our previous
observation that the herbicide activity of this class of compounds is
related to ability to block electron flow at the reducing side of PSII.
Additionally, our results are in agreement with a previous study
that showed that the ability of nostoclide analogues (another class
of butenolides that is structurally similar to rubrolides) to act as
photosynthesis inhibitors depends on their hydrosolubility. All of
the compounds that showed ability to interfere with the basal,
uncoupled and phosphorylating electron transport were investi-
gated to determine their site of action on the thylakoid electron
transport chain. Rubrolides analogues inhibit PSII by displacing
18] F. Bellina, C. Anselmi, V. Stéphane, L. Mannina, R. Rossi, Selective synthesis of
(Z)-4-aryl-5-[1-(aryl)methylidene]-3-bromo-2-(5H)-furanones, Tetrahedron
57 (2001) 9997–10007.
[
19] J.D. Mills, P. Mitchell, P. Schurrmann, Modulation of coupling ATPase activity in
intact chloroplasts, FEBS Lett. 191 (1980) 144–148.
[20] T.A.M. Veiga, B. King-Díaz, A.S.F. Marques, O.M. Sampaio, P.C. Vieira, M.F.G.F.
Silva, B. Lotina-Hennsen, Furoquinoline alkaloids isolated from
Balfourodendronriedelianum as photosynthetic inhibitors in spinach
chloroplasts, J. Photochem. Photobiol., B 120 (2013) 36–43.
[21] H.H. Strain, T. Cope, M.A. Svec, Analytical procedures for the isolation,
identification, estimation and investigation of the chlorophylls, Methods
Enzymol. 23 (1971) 452–466.
22] I. Yruela, G. Montoya, P.J. Alonso, R. Picorel, Identification of the pheophytin-
QA-Fe domain of the reducing side of the photosystem II as the Cu(II)-
inhibitory binding site, J. Biol. Chem. 266 (1991) 22847–22850.
[
[
23] R.T. Guiaquinta, R.A. Dilley, A partial reaction in photosystem II: reduction of
silicomolybdate prior to the site of dichlorophenyl dimethyl lurea inhibition,
Biochim. Biophys. Acta 387 (1975) 288–305.
B 1
Q at the D protein site, and have no effect on PSI, as demonstrat-
ed by both polarographic techniques and chlorophyll a fluores-
cence transients. Additionally, we found that all of the active
rubrolide analogues follow Tice’s rule of 5, which indicates that
these compounds may present physicochemical properties suit-
able for herbicides.
[24] B.R. Velthuys, Electron-dependent competition between plastoquinone and
inhibitors for binding to Photosystem II, FEBS Lett. 126 (1981) 277–281.
[
25] J.F. Allen, N.G. Holmes, in: M.F. Hipkinns, N.R. Baker (Eds.), Photosynthesis,
Energy Transduction. A Practical Approach, I.R.L. Press, Oxford, 1986, pp. 103–
141.
[
26] H.-X. Jiang, L.-S. Chen, J.-G. Zheng, S. Han, N. Tang, B.R. Smith, Aluminium
induced effects of photosystem II photochemistry in Citrus leaves assessed
by the chlorophyll a fluorescence transient, Tree Physiol. 28 (2008) 1863–
1871.
Acknowledgments
[
27] R.J. Strasser, M. Tsimilli-Michael, A. Srivastava, in: G.C. Papageorgion,
Govindjee (Eds.), Chlorophyll Fluorescence: A Signature of Photosynthesis,
Kluwer Academic Publishers, Netherlands, 2004, pp. 321–362.
[28] J. Boukouvalas, L.C. McCann, Synthesis of the human aldose redutase inhibitor
rubrolide L, Tetrahedron Lett. 51 (2010) 4636–4639.
29] M.L. Macías-Rubalcava, M.E.R.-V. Sobrino, C. Meléndez-González, B. King-Díaz,
B. Lotina-Hennsen, Selected phytotoxins and organic extracts from endophytic
fungus Edeniagomezpompae as light reaction of photosynthesis inhibitors, J.
Photochem. Photobiol., B 138 (2014) 7–26.
30] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Adv. Drug Deliv. Rev. 23 (1997) 3–25.
31] C.M. Tice, Selecting the right compounds for screening: does Lipinski’s Rule of
We are grateful to the following Brazilian agencies: Conselho
Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
[
(
CAPES) for research fellowships (EVVV, JOSV, and LCAB) and Fun-
dação de Amparo à Pesquisa de Minas Gerais (FAPEMIG) for sup-
porting our research in this area. The authors also gratefully
acknowledge the financial support from DGAPA-UNAM grants
PAPIT number IT102012-3 and PAIP number 4290-03 from the Fac-
ulty of Chemistry of UNAM.
[
[
5
1
for pharmaceuticals apply to agrochemicals?, Pest Manag Sci. 57 (2001) 3–
6.
[
[
32] MolinspirationCheminformatics, Bratislava, Slovak Republic, <http://
www.molinspiration.com>.
33] R.J. Strasser, A. Srivastava, Govindjee. Polyphasic chlorophyll a fluorescence
transients in plants and cyanobacteria, Photochem. Photobiol. 61 (1995) 32–
References
[
[
1] E.C. Oerke, Crop losses to pests, J. Agric. Sci. 144 (2006) 31–43.
2] P. Neve, M. Vila-Aiub, F. Roux, Evolutionary-thinking in agricultural weed
management, New Phytol. 184 (2009) 783–793.
42.

1
8
J.O.S. Varejão et al. / Journal of Photochemistry and Photobiology B: Biology 145 (2015) 11–18
[
[
34] M. Kitajima, W.L. Butler, Quenching of chlorophyll fluorescence and primary
photochemistry in chloroplasts by dibromothymoquinone, Biochim. Biophys.
Acta 376 (1975) 105–115.
35] S.Z. Tóth, G. Schansker, R.J. Strasser, In intact leaves, the maximum
fluorescence level (FM) is independent of the redox state of the
plastoquinone pool: A DCMU-inhibition study, Biochi. Biophys. Acta 1708
[36] L.N.M. Duysens, H.E. Sweers, Mechanisms of two photochemical reactions in
algae as studied by means of fluorescence, studies on microalgae and
photosynthetic bacteria. Special Issue of Plant and Cell Physiology, Japanese
Society of Plant Physiologists, University of Tokyo Press, considerable
simplification of the Tokyo, Japan, 1963, pp. 353–372.
(
2005) 275–282.