Synthesis and evaluation of antioxidant phenolic diaryl hydrazones as potent antiangiogenic agents in atherosclerosis

Article abstract of DOI:10.1016/j.bmc.2016.05.067

A series of bis-hydrazones derived from diaryl and diaryl ether hydroxybenzaldehyde frames 1 and 2 have been synthesized as potential antioxidant and antiangiogenic agents, two properties required to limit atherogenesis and cardiovascular events. These compounds were evaluated for their ability to neutralize free radical formation, to block endothelial cell-induced low-density lipoprotein oxidation (monitored by the formation of TBARS), an essential step in atherogenesis, and subsequent toxicity, to prevent angiogenesis evoked by low oxidized LDL concentration (monitored by the formation of capillary tubes on Matrigel) and to inhibit intracellular ROS increase involved in the angiogenic signaling. A structure/activity study has been carried out and finally allowed to select the phenolic diaryl ether hydralazine derivative 2a, sharing all these protective properties, as a promising hit for further development.

Full text of DOI:10.1016/j.bmc.2016.05.067

Accepted Manuscript
Synthesis and Evaluation of Antioxidant Phenolic Diaryl Hydrazones as Potent
Antiangiogenic Agents in Atherosclerosis
Corinne Vanucci-Bacqué, Caroline Camare, Chantal Carayon, Corinne Bernis,
Michel Baltas, Anne Nègre-Salvayre, Florence Bedos-Belval
PII:
S0968-0896(16)30411-4
DOI:
http://dx.doi.org/10.1016/j.bmc.2016.05.067
BMC 13051
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
11 April 2016
25 May 2016
29 May 2016
Please cite this article as: Vanucci-Bacqué, C., Camare, C., Carayon, C., Bernis, C., Baltas, M., Nègre-Salvayre,
A., Bedos-Belval, F., Synthesis and Evaluation of Antioxidant Phenolic Diaryl Hydrazones as Potent Antiangiogenic
Agents in Atherosclerosis, Bioorganic & Medicinal Chemistry (2016), doi: http://dx.doi.org/10.1016/j.bmc.
2016.05.067
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis and Evaluation of Antioxidant Phenolic Diaryl Hydrazones as Potent
Antiangiogenic Agents in Atherosclerosis
Corinne Vanucci-Bacqué,a,b Caroline Camare,c Chantal Carayon,a,b Corinne Bernis,c Michel
Baltas,a,b* Anne Nègre-Salvayre,c,* and Florence Bedos-Belval.a,b*
a Université Paul Sabatier Toulouse III; UMR 5068, Laboratoire de Synthèse et Physico-Chimie
de Molécules d’Intérêt Biologique ; 118, Route de Narbonne, F-31062 Toulouse Cedex 9,
France.
b
CNRS; UMR 5068, Laboratoire de Synthèse et Physico-Chimie de Molécules d’Intérêt
Biologique ; 118, Route de Narbonne, F-31062 Toulouse Cedex 9, France.
c INSERM; UMR1048, I2MC, BP 84225, 31432 Toulouse Cedex 4, France
* Corresponding author: Tel.: +33 (0)5 61 55 68 00 ; fax : +33 (0)5 61 55 60 11
E-mail address:
bedos@chimie.ups-tlse.fr;
baltas@chimie.ups-tlse.fr;
Anne.Negre-
salvayre@inserm.fr

Abstract
A series of bis-hydrazones derived from diaryl and diaryl ether hydroxybenzaldehyde frames 1
and 2 have been synthesized as potential antioxidant and antiangiogenic agents, two properties
required to limit atherogenesis and cardiovascular events. These compounds were evaluated for
their ability to neutralize free radical formation, to block endothelial cell-induced low-density
lipoprotein oxidation (monitored by the formation of TBARS), an essential step in atherogenesis,
and subsequent toxicity, to prevent angiogenesis evoked by low oxidized LDL concentration
(monitored by the formation of capillary tubes on Matrigel) and to inhibit intracellular ROS
increase involved in the angiogenic signaling. A structure/activity study has been carried out and
finally allowed to select the phenolic diaryl ether hydralazine derivative 2a, sharing all these
protective properties, as a promising hit for further development.
Keywords: Angiogenesis, antioxidant, atherosclerosis, biphenyl derivatives, polyphenol,
hydrazone.
1. Introduction

Atherosclerosis is a multifactorial and inflammatory pathology of large and medium arteries that
develops slowly and silently for years in the vascular wall, and may result in acute
cardiovascular events (infarction, stroke) due to plaque rupture and thrombosis.1 Among the
pathophysiological mechanisms involved in the development of atherosclerotic lesions and their
evolution toward advanced stages, the neovascularization of the plaque contributes to its
progression by providing nutrients, oxygen and blood cells. Moreover neovascularization
increases the risk of plaque instability and rupture, as it is correlated with local inflammation and
intraplaque microhemorrhages resulting from neocapillary apoptosis.2,3
Many factors present in atherosclerotic lesions may promote angiogenesis, including vascular
endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), transforming growth
factor β (TGFβ), lipid mediators such as lysophospholipids (sphingosine-1-phosphate or S1P,
lysophosphatidic acid or LPA), ischemia, hypoxia, oxidized lipids and oxidized low density
lipoprotein (oxLDL).4-8 All these factors are present in atherosclerotic plaques and contribute to
generate inflammatory events that are involved in the neovascularization process of the lesions.2-
5
Low density lipoprotein (LDL) are involved in atherogenesis after undergoing oxidation in the
intima.9 oxLDL exhibit a huge variety of proatherogenic properties including inflammatory,
migratory, mitogenic and apoptotic responses on vascular cells.10-12 The oxLDL implication in
angiogenesis has been recently reported in vitro and in vivo in the mouse Matrigel plug assay.13,14
The angiogenic signaling evoked by low oxLDL concentration depends on their content in
bioreactive oxidized lipids15,16 and on the subsequent generation of intracellular reactive oxygen
species (ROS), as supported by the inhibitory effect of antioxidants such as Trolox or N-
acetylcysteine (NAC).13,14,17 Interestingly, higher oxLDL concentration inhibit angiogenesis and
trigger apoptosis, suggesting that oxLDL may locally contribute to plaque neovascularization

and destabilization as well, through capillary apoptotic loss.14
Various classes of antioxidants including vitamins C and E and polyphenols, have been used for
inhibiting ROS generation, LDL oxidation and early atherogenic events in animal models for
atherosclerosis.18-21 However most antioxidants are poorly efficient against the late athero-
thrombotic events.22-24 Actually, there is a need of agents able to neutralize both ROS generation
and oxidized lipid bioreactivity, as they are the main trigger of oxLDL biological properties,
including angiogenesis. Ideally these compounds should share both antioxidant, carbonyl
scavenger and cytoptotective activities. For example, the antihypertensive drug hydralazine
neutralizes aldehydes generated from lipid oxidation and exerts antiatherogenic properties.25
However, hydralazine has poor antioxidant properties, and high concentrations are necessary for
reducing LDL oxidation by vascular cells by comparison with classical antioxidants.20,25,26
We recently developed a phenolic biphenyl bis-hydrazone prototype 1a (Fig. 1), exhibiting a
hydralazine moiety. This agent inhibits copper- and endothelial cell-induced LDL oxidation and
cytotoxicity, as well as the extra- and intra-cellular generation of ROS, and it exhibits
antiatherogenic properties in apoE-/- mice that spontaneously develop atherosclerotic lesions.27
In the present study, we have synthesized a series of phenolic bis-hydrazones based on prototype
1a, in order to perform a structure-activity comparison of the different agents to characterize and
improve their antioxidant and cytoprotective properties, and check whether they may slow down
the angiogenic process evoked by oxLDL. We initially chose to modulate the structure of 1a at
the hydrazone scaffold, which is known to induce various biological activities (namely anti-
inflammatory, cardio protective, anti-platelet and anti-cancer properties).28 Then, an ether link
between the two phenolic rings was introduced to figure out the influence of such a spacer. To
this aim, we prepared a variety of phenolic bis-hydrazones from two basic dialdehyde patterns,
the diaryl linked derivative 1, and the related diaryl ether 2 (Fig. 1).

Figure 1. Synthetic strategy scope - Chemical structure of compound 1a and dialdehyde
precursors 1 and 2.
We have compared the different properties exhibited by the new synthesized agents including
their antioxidant activity, either by using the physicochemical DPPH assay, or by testing their
ability to block LDL oxidation by endothelial cells (TBARS assay), and their ability to inhibit
the generation of intracellular ROS that are involved in the angiogenic process evoked by
oxLDL. Indeed, the source, the nature and the mechanism of ROS generation involved in LDL
oxidation or in angiogenesis may be different, and it is relevant to evaluate the capacity of the
agents to inhibit both intra- and extracellular ROS production. Based on these results, we tested
the ability of the different diaryl derivatives to block the formation of microtubes by
microvascular endothelial cells (HMEC-1) on Matrigel (a classical model for studying in vitro
angiogenesis), upon stimulation by oxLDL.
2. Results and Discussion
2.1. Chemistry
The 6,6’-dihydroxy-5,5’-dimethoxybiphenyl-3,3’-dicarbaldehyde (1) was prepared as previously
reported.29

Concerning the diaryl ether scaffold synthesis (Scheme 1), we previously demonstrated that
Ullmann standard coupling conditions (Cu(0) in refluxing DMF) applied to 4-benzyloxy-3-
hydroxy-5-methoxybenzaldehyde (3) and 4-benzyloxy-3-bromo-5-methoxybenzaldehyde (4) led
to the symmetric dibenzylated diaryl ether 5 as traces (8% yield). A benzyl migration in the
phenolic partner is implicated in the poor yield.30 So we envisioned the use of differently O-
protected reactants, with first focus on O-silylated compounds. Unfortunately, regioselective
TBS-silylation of 5-hydroxyvanillin failed. Alternatively, commercially available 3,4-
dimethoxy-5-hydroxybenzaldehyde (6) and 3-bromo-4,5-dimethoxybenzaldehyde (7) were
reacted in refluxing DMF with Cu(0) affording the expected symmetric diaryl ether 8 in high
yield (87%). The optimized regioselective demethylation of the 4-methoxy groups of
tetramethoxy ether 8 was achieved by treatment with AlCl3 (6 equiv.) in dichloromethane,
cleanly providing the expected diphenol 2 in 81% yield (Scheme 1).
Scheme 1. Synthesis of the diaryl ether core. Reagents and conditions: a) Cu(0), DMF, reflux;
b) AlCl3, CH2Cl2, rt.
With the required dialdehydes in hand, we focused on the introduction of the hydrazone
functionalities. As shown in Scheme 2, hydrazone-type derivatives were efficiently prepared by
condensation of two strict equivalents of various commercially available hydrazines/hydrazides

(1-hydrazinylphthalazine hydrochloride, isonicotinic hydrazide, 3-methyl-1H-pyrazole-5-
carbohydrazide, 2-hydrazinylbenzo[d]thiazole and benzyl hydrazinecarboxylate) on dialdehydes
1 and 2 in refluxing ethanol (75-99% yields).
Scheme 2. Synthesis of bis-hydrazones. Reagents and conditions: a) 1-hydrazinylphthalazine
hydrochloride (2 equiv.), EtOH, reflux; b) isonicotinic hydrazide (2 equiv.), EtOH, reflux; c) 3-
methyl-1H-pyrazole-5-carbohydrazide (2 equiv.), EtOH, reflux; d) 2-hydrazinylbenzo[d]thiazole
(2 equiv.), EtOH, reflux; e) benzyl hydrazinecarboxylate (2 equiv.), EtOH, reflux.
2.2. Antiradical activities
Radical scavenging potency of the tested compounds was assessed in vitro by the DPPH assay,
mostly used for phenol derivatives.

Table 1. Free radical scavenging activity (DPPH), inhibition (TBARS) of cell-induced LDL
oxidation and cytoprotective effect (MTT) against oxLDL toxicity of the diaryl derivatives at 1
µM and 10 µM.
Compound
DPPH
TBARS (%)c
MTT (%)d
10µM
IC50 (µM)a
382.1 13
9.7 0.3
1µM
10µM
88 9
8 4
1µM
19 4
52 5
18 4
14 2
58 6
52 4
32 2
56 6
42 5
41 5
51 2
51 4
103 5
56 4
21 3
99 4
19 5
40 7
91 6
90 9
35 6
72 9
66 4
65 6
69 4
69 4
1
1a
1b
1c
1d
1e
2
38.8 0.3
29.1 0.3
27.7 0.3
59.7 1.4
289.4 6.0
11.5 0.3
21.5 0.9
19.8 2.5
8.6 0.5b
23.6 0.1
20.1 1.4
(17.7 0.4)b
112 10
99 3
92 8
71 12
7 2
19 9
18 4
5 1
95 10
37 7
87 22
10 3
88 15
91 20
12 5
6 2
2a
2b
2c
2d
2e
100 2
97 4
93 12
28 7
Trolox
57 2
10 6
45 5
20 5
68 12
100 5
Control
a Determined by UV-absorption at 517 nm. Results are the means SEM (n=3)

b Determined by EPR.
c
Results are expressed as percent of the control incubated with LDL and HMEC-1 in the
absence of agent. The data are mean SEM of 4 separate experiments. p < 0.05.
d The results are expressed as % of the untreated control. Control means with oxLDL without
agent.
The DPPH assay measures the reduction in absorbance of a free stable 2,2′-diphenyl-1-
picrylhydrazyl (DPPH) radical ethanolic solution at 517 nm,31 in the presence of the tested
derivatives. It is noted that in the case of benzothiazol derivative 2d, for which a staining
appeared in the presence of the DPPH radical, preventing the reading of the absorbance at
517nm, the measurement was determined using EPR spectroscopy.32,33 Table 1 shows the
concentration for 50% of inhibition of DPPH free radical. Trolox (6-hydroxy-2,5,7,8-
tetramethylchroman-2-carboxylic acid) known for its antioxidant capacity, a water-soluble
vitamin E analog, was used as standard reference.
It is noteworthy that the phenolic dialdehydes 1 and 2 show no DPPH radical scavenger activity
(IC50 values 382.1 13.3 µM and 289.4 6.0 µM respectively). The phenolic moiety is thus not
enough in itself to ensure efficient antiradical ability of these compounds. The hydrazone frame
is crucial for the antioxidant capacity of studied compounds due to the π-electron delocalization
increase.34 Indeed it is observed that compounds 1a, 2a and 2d exhibit antiradical properties
twice more effective than Trolox, while 2b, 2c and 2e displays IC50 value in the same range as
Trolox (Table 1). All the other hydrazones 1b-e exhibit DPPH scavenging activities (27.7 0.3
≤IC50≤ 59.7 1.4 µM) positioning them as good radical trapping agents in the DPPH assay
compared to the non-antiradical dialdehydes 1 and 2. In most cases, the phenolic diaryl ether

derivatives show better scavenging capacities than the corresponding diaryl ones. However, the
DPPH assay gives only an indication of the antioxidant potential of a molecule, as it is carried
out in the absence of cells.
2.3. Antioxidant activity: Protective effect against cell-induced LDL oxidation
ROS produced by vascular cells, play a major role in atherogenesis and in the formation of early
atherosclerotic lesions, by triggering LDL oxidation in the intima.9,18,35 This mechanism can be
mimicked in vitro by incubating vascular cells with LDL.25,27 The LDL oxidation process occurs
°-
extracellularly and involves the superoxide anion O2 that is secreted upon stimulation of
NADP(H) oxidase (NOX).36 Antioxidants may inhibit LDL oxidation either via the
neutralization of ROS in the extracellular medium, or/and by inhibiting NOX activation.18-23 We
checked whether the newly synthesized hydrazones inhibit the oxidation of native LDL by
human endothelial cells (HMEC-1).
LDL oxidation by HMEC-1 was determined in the absence (control), or in the presence of each
agent (tested at 1 and 10 µM final concentration), and was assessed by measuring the generation
of thiobarbituric acid reactive substance (TBARS) in the LDL-containing culture medium.37 The
results were compared to those obtained with LDL without any agent in presence of the vehicle
(DMSO, 0.1% final concentration). The data presented in Table 1 are expressed as percent of
TBARS generated in the antioxidant-containing culture medium vs TBARS generated in the
absence of antioxidant (100%), using Trolox as reference.
These data point out the different efficacy of the synthesized derivatives. First of all, it should be
noted that the phenolic dialdehyde precursors 1 and 2 do not exhibit any antioxidant activity by
themselves, meaning that other functionalities are required to induce this property, as also

observed in the DPPH assay. Besides, all the tested agents display a better antioxidant activity
when tested at 10 µM rather than 1 µM, justifying their use in further experiments at 10 µM.
The diaryl phenols 1a, 1d, 1e, 2a, 2d and 2e (TBARS values from 5 to 12%) were as efficient as
Trolox (10%) at 10 µM. Interestingly, when used at lower concentration (1µM), compounds 1d,
1e, 2a and 2e, were more efficient than Trolox (1 µM) to block LDL oxidation (TBARS values
ranging from 18 to 37%, vs Trolox, 57%).
The other agents were either poorly or not efficient. In particular, compounds 1b-2b and 1c-2c
exhibiting isonicotinoyl hydrazide or 3-methyl-1H-pyrazole-5-carbohydrazide moieties
respectively, were surprisingly not active as antioxidants, in spite of their phenolic moieties and
of good radical scavenging properties (Table 1).
When comparing diaryl and corresponding diaryl ether frames for a same functionalization, it is
to note that the oxygen spacer does not play any key role in the ability to block cell-induced LDL
oxidation, whereas diaryl ether structures exhibit a better radical scavenging activity (Table 1).
This suggests that difference in antioxidant efficacy could result from their ability to neutralize
free radicals produced by endothelial cells, but also to modulate the activity of cellular enzymatic
systems and signaling pathways involved in free radical production.
Since high concentrations of oxidized LDL are toxic for vascular cells,38,39 we checked whether
the newly synthesized hydrazones may inhibit this toxicity. The toxicity of oxidized LDL (200
µg mL-1) was studied on HMEC-1, either with the vehicle only (DMSO 0.1%), or in the presence
of hydrazones (1 and 10 µM), using Trolox as control. The cell viability was studied using the
MTT test 27 and is reported in Table 1.
We first tested the self-toxicity of the different newly synthesized hydrazones, in the absence of
oxLDL. This was studied by checking whether the different agents added to the culture medium

at 10 µM in the vehicle (0.1%), trigger a loss of cell viability (detected by the MTT assay). In the
used conditions, no toxicity was observed for any compounds at 1 and 10 µM concentrations
(data in Supporting Information).
Regarding their cytoprotective effect, dialdehydes 1 and 2 were found ineffective. Most diaryl
hydrazones inhibited oxLDL toxicity, particularly compounds 1a, 1d and 1e exhibiting 90 to
99% of cell viability at 10 µM, higher than Trolox (68%), knowing that the residual viability of
HMEC-1, incubated for 36 h with toxic oxLDL, was less than 20%. The latter derivatives were
as efficient as Trolox at 1 µM. Furthermore diaryl compounds 1b and 1c have neither
cytoprotective effect nor antioxidant activity in this system, which suggests that these agents are
either inactive against free radicals generated by endothelial cells, or are unable to counteract the
apoptotic signaling of oxLDL .
The protective effect of phenol diaryl ether series 2a-e was as efficient as Trolox in preventing
oxLDL toxicity at 1 and 10 µM. It is noteworthy that compounds 2b and 2c were cytoprotective
though they exert a poor antioxidant activity, as they did not protect LDL from oxidation (Table
1). These results suggest that compounds 2b and 2c could exert their cytoprotective activity by
inhibiting the toxic and proapoptotic cell signaling evoked by oxLDL,40 however it is likely that
they should not slow down the accumulation of foam cells and fatty streaks which are directly
related to the uptake of oxLDL by macrophages,13,41 nor angiogenesis which depends on ROS
generation.13,14 Anyhow, our results indicate that the deleterious signaling evoked by oxLDL,
oxidized lipids and oxidative stress, may be inhibited either by antioxidants (which also exhibit
cytoprotective activity), or by cytoprotective agents devoid of any antioxidant activity.
2.4. Inhibition of capillary formation on Matrigel by HMEC-1, induced by oxidized LDL

We then checked whether the synthetized phenolic diaryl derivatives could inhibit the
angiogenicity of oxLDL evidenced by the formation of capillary tubes on Matrigel by HMEC-1,
in the presence of low oxLDL concentration (20 µg mL-1).35 Diaryl hydrazones 1a, 1d, 1e and
corresponding diaryl ether hydrazones 2a, 2d and 2e were selected as the most efficient
compounds against cell-induced LDL oxidation at 10 µM (Table 1). Trolox was used as
antioxidant and antiangiogenic control.14
Figure 2. Capillary tube formation by HMEC-1 stimulated by low oxLDL concentration (20 µg
mL-1) w/wo diaryl phenol derivatives and Trolox (10 µM each). A) Quantification of formed
capillary tubes by HMEC-1 grown on Matrigel. Tube formation was expressed as % of linked
cells to the total cell number. Mean
SEM of
4
separate experiments,
#(oxLDL/control),*(agent/oxLDL) p<0.05. FBS, fetal bovine serum. B) Representative pictures
of tube formation on Matrigel in the presence of the different agents.
We recently reported that low oxLDL concentration (20-50 µg/apoB/mL) elicit the formation of
capillary tubes by HMEC-1 on Matrigel.14,17 We chose an angiogenic concentration (20 µg/mL)

for the anti-angiogenic effect evaluation of the synthesized agents. As shown in Fig. 2, low
oxLDL concentration (20 µg mL-1) stimulated the formation of capillary tubes by HMEC-1, in
the same range than the positive control consisting in fetal bovine serum (FBS, 2.5%). The
phenolic diaryl derivatives 1a, 1d and 1e display a similar strong efficiency in inhibiting the
capillary tube formation by HMEC-1 in the same range as Trolox, whatever their hydrazone
moiety nature. On the other hand, the efficiency of the tested phenolic diaryl ether analogues
depends on the nature of the hydrazone part: benzothiazolhydrazone 2d exihibits a similar
antiangiogenic activity as Trolox, whereas benzyl carboxylate hydrazone 2e is totally inefficient
though this agent efficiently inhibited LDL oxidation (Table 1). Gratefully, hydralazine diaryl
ether 2a is much more potent than Trolox (less than 5% of residual tube vs 20% respectively).
Noteworthy, in this case the ether link promotes the antiangiogenic activity since 2a is four times
more potent than the corresponding diaryl analogue 1a.
As the angiogenic signaling of oxLDL involves a rapid generation of intracellular ROS,14,17 we
checked whether the protective antiangiogenic effect of the selected derivatives is correlated with
an inhibition of this intracellular ROS increase in HMEC-1 (detected using the ROS-sensitive
probe H2DCFDA-AM).

Figure 3. Effect of phenolic diaryl derivatives on the intracellular ROS increase evoked by
oxLDL. ROS were determined in HMEC-1 preloaded with the fluorescent H2DCFDA-AM
probe (5 µM), after 40 min incubation with oxLDL (20 µg mL-1), w/wo phenolic diaryl
derivatives or Trolox (10 µM each). Mean
#(oxLDL/control),*(agent/oxLDL) p<0.05.
SEM of 4 separate experiments,
As shown in Fig. 3, low angiogenic oxLDL concentration rapidly stimulates the generation of
intracellular ROS. This ROS increase involves the oxLDL receptor LOX-1 and the activation of
NOX2.13,17 We show here that most tested agents inhibit the intracellular ROS increase, which
suggests that they may either directly inhibit NOX activation at the plasma membrane or block
the intracellular signaling leading to NOX activation and ROS production. Interestingly, the

diarylether 2e was unable to block the intracellular ROS increase evoked by oxLDL, this being
correlated to its lack of inhibitory effect against tube formation on Matrigel (Fig.2), whereas this
agent 2e efficiently blocked LDL oxidation (Table 1). These data suggest that 2e is able to
neutralize ROS produced in the extracellular medium, but it is probably unable to block the intra-
cell signaling involved in ROS production. These results, together with their potent
antiangiogenic effect confer an originality to molecules 1a, 1e, 1d, 2a and 2d against oxLDL
atherogenic signaling.
3. Conclusion
In conclusion, different phenolic diaryl and diaryl ether derivatives featuring hydrazone
functionalities have been synthesized, and their antioxidant and antiangiogenic properties have
been evaluated and compared. This study indicates that despite their well-established antioxidant
properties, phenol scaffolds are not sufficient to induce this required dual activity, which is
mainly modulated by the hydrazone functionalization.
The biological data observed with the phenolic hydrazone diaryls 1a, 1d, 1e and diaryl ethers 2a,
2d as promising agents show that they are able to block LDL oxidation elicited by vascular cells,
the toxic effect of high oxLDL concentrations for vascular cells, the intracellular ROS increase
and the subsequent tube formation characteristic of the proangiogenic oxLDL properties.
Particularly, the striking antiangiogenic efficiency of the phenolic hydralazine diaryl ether 2a
allows to propose this component as a promising candidate for future studies.
4. Experimental Section
4.1. Chemistry

Melting points (mp) were obtained on a Buchi apparatus and are uncorrected. Infrared (IR)
spectra were recorded on a Thermo Nicolet Nexus spectrometer. Nuclear magnetic resonance
(NMR) spectra were recorded on Brucker Avance 300 MHz spectrometers. Chemical shifts (δ)
1
were reported in parts per million (ppm) relative to TMS at δ = 0 ppm for H NMR spectra and
to residual solvent peak signals for 13C NMR spectra. Signals are described as follow: s, singlet;
br, broad signal; d, doublet; t, triplet; m, multiplet. High resolution mass spectra (HRMS) were
recorded on a Xevo G2 QTOF (Waters) instrument.
Compounds 1, 1a, 1b29 and 1d42 have been previously reported and fully characterized.
4.1.1. 5,5'-Oxybis(3,4-dimethoxybenzaldehyde) (8).
A
mixture of 3-hydroxy-4,5-
dimethoxybenzaldehyde (6) (576 mg, 3.17 mmol), 3-bromo-4,5-dimethoxybenzaldehyde (7)
(2.33 g, 9.51 mmol ) and copper powder (1 g, 15.85 mmol) in dry DMF (6 mL) was heated at
reflux temperature under argon atmosphere for 16 h. The reaction mixture was cooled to room
temperature, diluted with EtOAc and filtrated over a Celite® pad. Water was added and the
aqueous layer was extracted with EtOAc. The combined organic layer was washed with brine,
dried over MgSO4, filtrated and concentrated under vacuum. Purification by silica gel column
chromatography (EtOAc/Petroleum ether : 20/80) allowed compound 8 (958 mg, 87%) as a
1
colorless solid. mp : 94-96 °C; IR (neat): υ max = 1693 cm-1; H NMR (300 MHz, CDCl3,): δ =
13
3.96 (s, 6H), 3.97 (s, 6H), 7.03 (d, J = 1.8 Hz, 2H), 7.28 (d, J = 1.8 Hz, 2H), 9.79 (s, 2H); C
NMR (75 MHz, CDCl3): δ = 56.3, 61.1, 107.5, 114.5, 131.7, 145.4, 149.9, 154.3, 190.4; HRMS
(DCI, CH4) : m/z [M + H]+calcd for C18H19O7: 347.1131, found: 347.1135.
4.1.2. 5,5'-Oxybis(4-hydroxy-3-methoxybenzaldehyde) (2). To a stirred solution of compound
8 (662 mg, 1.91 mmol) in dry CH2Cl2 (33 mL) was added AlCl3 (1.53 g, 11.5 mmol). The
reaction mixture was stirred at room temperature under argon atmosphere for 6 h. Water (24 mL)

was slowly added and the mixture was filtered over a sintered glass funnel. The aqueous layer
was extrated with CH2Cl2. EtOAc (30 mL) was added to the collected precipitate. The resulting
mixture was stirred at 60 °C for 15 min then cooled to room temperature. After filtration, the
filtrate was combined to the CH2Cl2 organic extracts. The organic layers were dried over MgSO4,
filtrated and concentrated under reduced pressure. Silica gel column chromatography
(EtOAc/petroleum ether: 90/10) afforded pure compound 2 (493 mg, 81%) as a beige solid.
Characterization data were identical with that previously reported.30
4.2. General procedure for the synthesis of the dihydrazone-type compounds.
A mixture of the appropriate hydrazine or hydrazide (2 equiv.) and dialdehyde (1 equiv.) was
dissolved in absolute EtOH (30 mL/mmol) and heated under reflux for 18 h. After cooling to
room temperature, the precipitated solid was collected by filtration and dried under vacuum to
afford the corresponding bis-hydrazone. If no precipitate was observed, the reaction mixture was
concentrated under reduced pressure.
4.2.1.
(N',N''E,N',N''E)-N',N''-(6,6'-Dihydroxy-5,5'-dimethoxybiphenyl-3,3'-
diyl)bis(methan-1-yl-1-ylidene)bis(3-methyl-1H-pyrazole-5-carbohydrazide) (1c). Following
the general procedure, starting with 1 (50 mg, 0.165 mmol) and 1-hydrazinylphthalazine
hydrochloride (46 mg, 0.33 mmol) in ethanol, compound 1c was obtained as a yellow solid (87
1
mg, 97 % yield). mp >260 °C; IR (neat): υmax = 3426, 3224, 1661, 1602 cm-1; H NMR (300
MHz, (CD3)2SO): δ = 2.28 (s, 6H), 3.91 (s, 6H), 6.48 (s, 2H), 7.01 (d, J = 1.4 Hz, 2H), 7.30 (d, J
13
= 1.4 Hz, 2H), 8.39 (s, 2H), 8.92 (br s, 2H), 11.40 (s, 2H), 13.04 (s, 2H); C NMR (75 MHz,
(CD3)2SO): δ = 10.4, 56.0, 104.8, 107.3, 124.3, 125.1, 140.0, 146.0, 146.2, 147.9, 148.2, 158.3;
HRMS (ESI+) m/z [M+Na]+ calcd for C26H26N8O6Na: 569.1873, found: 569.1900.

4.2.2. (2E,2'E)-Benzyl 2,2'-(6,6'-dihydroxy-5,5'-dimethoxybiphenyl-3,3'-diyl)bis(methan-1-
yl-1-ylidene)bis(hydrazinecarboxylate) (1e). Following the general procedure, starting with 1
(50 mg, 0.165 mmol) and benzyl hydrazinecarboxylate (55 mg, 0.33 mmol) in ethanol,
compound 1e was obtained as a yellow solid (94 mg, 95 % yield). mp : 135-137 °C; IR (neat):
1
υmax = 3440, 1718, 1622 cm-1; H NMR (300 MHz, (CD3)2SO): δ = 3.88 (s, 6H), 5.16 (s, 4H),
6.97 (d, J = 1.8 Hz, 2H), 7.21 (d, J = 1.8 Hz, 2H), 7.37-7.41 (m, 10H), 7.95 (s, 2H), 8.92 (br s,
2H), 11.11 (br s, 2H); 13C NMR (75 MHz, (CD3)2SO): δ = 55.9, 65.7, 107.4, 123.4, 124.8, 125.1,
127.9, 128.0, 128.4, 136.7, 144.9, 145.8, 148.0, 153.4; HRMS (ESI+) m/z [M+Na]+ calcd for
C32H30N4O8Na: 621.1961, found: 621.1960.
4.2.3.
(E)-6,6'-Oxybis(2-methoxy-4-((E)-(2-(phthalazin-1-yl)hydrazono)methyl)phenol)
dihydrochloride (2a). Following the general procedure, starting with 2 (70 mg, 0.22 mmol) and
1-hydrazinylphthalazine hydrochloride (86.5 mg, 0.44 mmol) in ethanol, compound 2a was
obtained as a yellow solid (112 mg, 76 % yield). mp >280°C; IR (neat): υmax = 3361, 3204,
1
1617, 1595 cm-1; H NMR (300 MHz, (CD3)2SO): δ = 3.99 (s, 6H), 7.09 (d, J = 1.7 Hz, 2H),
7.70 (d, J = 1.7 Hz, 2H), 8.09-8.19 (m, 6H), 8.88 (br, 2H), 8.99 (br, 2H), 9.05-9.17 (m, 2H), 9.7
13
(br, 2H), 14.3 (br, 2H); C NMR (75 MHz, (CD3)2SO): δ = 56.4, 106.4, 113.5, 118.9, 123.2,
125.5, 127.8, 128.1, 133.6, 135.8, 141.1, 144.4, 144.6, 147.4, 149.4, 152.9; HRMS (ESI+) m/z [M
+ H] + calcd for C32H27N8O5: 603.2104, found: 603.2109.
4.2.4.
(N',N''E,N',N''E)-N',N''-(5,5'-Oxybis(4-hydroxy-3-methoxy-5,1-
phenylene))bis(methan-1-yl-1-ylidene)diisonicotinohydrazide (2b). Following the general
procedure, starting with 2 (119 mg, 0.37 mmol) and isonicotinic hydrazide (102.5 mg, 0.74
mmol) in ethanol, compound 2b was obtained as a white solid (199 mg, 96 % yield). mp : 230-

1
231°C; IR (neat): υmax = 3378, 3274,1652 cm-1; H NMR (300 MHz, (CD3)2SO): δ = 3.90 (s,
6H), 6.85 (d, J = 1.7 Hz, 2H), 7.14 (d, J = 1.7 Hz, 2H), 7.77 (dd, J = 1.6, 4.4 Hz, 4H), 8.30 (s,
2H), 8.76 (dd, J = 1.6, 4.4 Hz, 4H), 9.48 (br s, 2H), 11.89 (br, 2H); 13C NMR (75 MHz,
(CD3)2SO): δ = 56.6, 106.3, 110.7, 121.4, 124.4, 140.0, 140.5, 144.8, 149.0, 149.1, 150.2, 161.3;
HRMS (ESI+) m/z [M+Na] + calcd for C28H24N6O7Na: 579.1604, found: 579.1601.
4.2.5.
(N',N'',E,N',N'',E)-N',N''-(5,5'-Oxybis(4-hydroxy-3-methoxy-5,1-
(2c).
phenylene))bis(methan-1-yl-1-ylidene)bis(3-methyl-1H-pyrazole-5-carbohydrazide)
Following the general procedure, starting with 2 (119 mg, 0.37 mmol) and 3-methyl-1H-
pyrazole-5-carbohydrazide (105 mg, 0.74 mmol) in ethanol, compound 2c was obtained as a
white solid (190 mg, 90 % yield). mp >280°C; IR (neat): υmax = 3539, 3497, 3250, 1680, 1652
1
cm-1; H NMR (300 MHz, (CD3)2SO) δ = 2.26 (s, 6H), 3.89 (s, 6H), 6.45 (br s, 2H), 6.73 (br,
2H), 7.07 (br, 2H), 8.29 (s, 2H), 9.38 (br, 2H), 11.37 (s, 2H), 13.02 (s, 2H); 13C NMR (75 MHz,
(CD3)2SO) δ = 10.3, 56.1, 104.8, 105.6, 110.8, 125.0, 139.6, 140.0, 144.9, 146.1, 147.3, 149.2,
158.1; HRMS (ESI+) m/z [M+Na] + calcd for C26H26N8O7Na: 585.1822, found: 585.1823.
4.2.6. (E)-6,6'-Oxybis(4-((E)-(2-(benzo[d]thiazol-2-yl)hydrazono)methyl)-2-methoxyphenol)
(2d). Following the general procedure, starting with 2 (122 mg, 0.38 mmol) and 2-
hydrazinylbenzo[d]thiazole (127 mg, 0.76 mmol) in ethanol, compound 2d was obtained as a
1
white solid (195 mg, 83 % yield). mp >280°C; IR (neat): υmax = 3490, 1617, 1578 cm-1; H
NMR (300 MHz, (CD3)2SO): δ = 3.90 (s, 6H), 6.81 (d, J = 1.6 Hz, 2H), 7.04 (td, J = 7.6, 1.1 Hz,
2H), 7.14 (d, J = 1.6 Hz, 2H), 7.25 (td, J = 7.6, 1.1 Hz, 2H), 7.34-7.41 (m, 2H), 7.70 (d, J = 7.6
Hz, 2H), 7.97 (s, 2H), 9.37 (br s, 2H), 12.13 (br s, 2H); 13C NMR (75 MHz, (CD3)2SO): δ = 56.0,
101.1, 105.4, 110.2, 117.2, 121.3, 121.4, 124.9, 125.8, 129.2, 139.2, 143.9, 144.8, 149.1, 166.8;
HRMS (ESI+) m/z [M + H] + calcd for C30H25N6O5S2: 613.1328, found: 613.1329.

4.2.7. (2E, 2'E)-Benzyl 2,2'-(5,5'-Oxybis(4-hydroxy-3-methoxy-5,1-phenylene))bis(methan-
1-yl-1-ylidene)bis(hydrazinecarboxylate) (2e). Following the general procedure, starting with 2
(64 mg, 0.20 mmol) and benzyl hydrazinecarboxylate (66 mg, 0.40 mmol) in ethanol, compound
2e was obtained as a white solid (98 mg, 80 % yield). mp (EtOAc) : 133-135 °C; IR (neat): υmax
= 3381, 3211, 1699 cm-1; 1H NMR (300 MHz, CD3OD): δ = 3.93 (s, 6H), 5.20 (s, 4H), 6.70 (d,
J = 1.7 Hz, 2H), 7.32-7.39 (m, 12H), 7.74 (br, 2H); 13C NMR (75 MHz, CD3OD): δ = 56.8, 68.0,
105.8, 113.5, 126.7, 129.1, 129.2, 129.5, 137.8, 140.9, 145.9, 146.7, 150.7, 156.4; HRMS (ESI+)
m/z [M + H] + calcd for C32H31N4O9: 615.2091, found: 615.2087.
4.3. DPPH radical scavenging activity
To 4 mL of a DPPH• solution in ethanol (concentration 85.65 3.37 µM to obtain a DO = 1) was
added 5 to 50 µL of compounds solution in ethanol at different concentrations (4–800 µM). The
absorbance was measured at 517 nm, at t = 0 min (A0) with Spectrophotometer UV- Visible
Perkin-Elmer lambda 17 and after 60 min incubation at room temperature (A60). Percentage
inhibition of DPPH free radical (I%) was calculated in the following way: I% =
100×(A0−A60)/A0. IC50 represents concentration of the tested compound, providing 50%
inhibition of DPPH radicals. All experiments were carried out in triplicate. The 6-hydroxy-
2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) was used as antioxidant reference.
4.4. Cell Biology methods
4.4.1. Chemicals. Matrigel was from BD Biosciences (Le-Pont-de-Claix, FR). Corning
Transwell Polycarbonate Membrane 24-wells, calcein-AM, MTT, from Sigma-Aldrich.
H2DCFDA-acetoxymethyl ester was from Invitrogen, cell culture reagents and other reagents

were from WWR or Sigma.
4.4.2. Cell culture. Human microvascular endothelial cells (HMEC-1) were from the CDC (Dr.
Candal, Atlanta, US). HMEC-1 cell line is an immortalized human dermal microvascular
endothelial cell line that can be passaged more than 95 times without signs of senescence. These
cells express the von Willebrand's Factor specific for endothelial cells, CD31, CD36, adhesion
molecules, and form tubes when cultured on Matrigel, thus can be suitably utilized for
angiogenesis assay. HMEC-1 may oxidize human LDL, and are suitable for studying the
protective effect of antioxidants, as reported.25,27 In this study, HMEC-1 were used at passage
30-36. Cells were grown in MCDB-131 culture medium supplemented with 10% heat inactivated
fetal calf serum, 100 U mL-1 penicillin and 100 µg mL-1 streptomycin. The cells were starved in
serum-free RPMI medium 24h before the experiments.
4.4.3. LDL isolation and oxidation. LDL from human pooled sera were prepared by
ultracentrifugation, dialyzed against PBS containing 100 µmol.L-1 EDTA, as reported.27 Cell-
induced LDL oxidation and the protective effect of the diaryl derivatives were studied on
HMEC-1 seeded in 12-multiwell plates. Cells were incubated in serum-free RPMI 1640
containing native LDL (100 µg apoB mL-1). The synthesized compounds were solubilized in
DMSO, and added to the culture medium at the indicated concentrations (and 0.1 % DMSO final
concentration), simultaneously with LDL. At the end of the incubation (14 h at 37 °C), the LDL-
containing medium was used for determining the TBARS content,27 using Trolox as control
antioxidant. For angiogenic and toxicity experiments, LDL were mildly oxidized by UV in the
absence of the agent or Trolox as described.27 Under standard conditions, UV-oxLDL contained
71-104 nmol lipid hydroperoxide/mg apoB and 6.4-9.7 nmol TBARS/mg apoB.

4.4.4. Capillary tube formation. HMEC-1, seeded at 30,000 cells/well on Matrigel in 24
multiwell plates, were grown for 18 h at 37 °C, in 500 µL MCDB-131 medium, containing either
0.1% fetal bovine serum (FBS, negative control) or 2.5 % FBS (positive control) or oxLDL (20
µg mL-1), with or without phenolic diaryl or diaryl ether derivatives (10 µM). Capillary tubes
were stained by Calcein (1 µM, 30 min), and photographed (Nikon Coolpix 995 camera) under a
fluorescence microscope (λex 496, λem 516 nm). The number of linked cells, (i.e. cells
connected to each other by thin microtube network) was counted and reported to the total cell
number (linked + not-linked cells).14,17
4.4.5. Intracellular ROS determination. ROS generated in cells treated by oxLDL (20 µg mL-
1) were evaluated by measuring the oxidation of 2’, 7’-dichlorodihydrofluorescein diacetate
acetoxymethyl ester (H2DCFDA-AM) at the indicated times.14 The probe was added to the
culture medium (5 µM final concentration) 30 min before the end of the experiment. Then, the
cells were washed three times with PBS. The fluorescence of the cell homogenate was measured
(λex 495, λem 520 nm). The data are expressed as ratio of fluorescence/ fluorescence of the
unstimulated control.14 Note that ROS determination could not be performed on Matrigel,
because of the adsorption of the probe on this medium.
4.4.6. Evaluation of cytoprotective effect. LDL were mildly oxidized by UV-C irradiation.14,17
Briefly, LDL (2 mg apoB L-1 in dialysis buffer) were irradiated for 1 h and 30 min as a thin film
(5 mm) in an open beaker placed 10 cm under the UV-C source (HNS 30W OFR Osram UV-C
tube, λmax 254 nm, 0.5 mW cm-2). This method produces mildly UV-oxidized LDL (here
referred to as oxLDL) that contained 63-97 nmol lipid hydroperoxide/mg apoB 3, and 5.6-8.3
nmol TBARS/mg apoB. OxLDL were sterilized by filtration and immediately used. For
evaluating the protective effect of the synthesized antioxidants, HMEC-1 were seeded in 12

multiwell culture plates, in DMEM supplemented with 10% FBS. At sub-confluency, this
medium was discarded and replaced by fresh RPMI medium containing 200 µg mL-1 of UV-
oxidized LDL, and the different agents at the indicated concentration. After 24 h of contact, the
whole cytotoxicity was evaluated by the MTT assay. A control without LDL was done in the
same conditions. The results were expressed as percent of the unstimulated control (100%
viability).
4.4.7. Statistical analysis. Data are given as mean SEM. Estimates of statistical significance
were performed by analysis of variance (one way Anova, Tukey test, SigmaStat software).20,27
Acknowledgements
The authors acknowledge the financial support from INSERM, CNRS, Université Paul Sabatier,
ANR Carina, Synfomia project and the technical support from Institut de Chimie de Toulouse
(FR 3599).
Supplementary data
Supplementary data (NMR spectra and self-toxicity) associated with this article can be found, in
the online version, at
References and notes
1. Hopkins, P. N. Physiol. Rev. 2013, 93, 1317-542.
2. Carmeliet, P.; Jain, R. K. Nature 2011, 473, 298-307.
3. Michel, J. B.; Virmani, R.; Arbustini, E.; Pasterkamp, G. Eur. Heart J. 2011, 32, 1977-1985.
4. Mulligan-Kehoe, M. J.; Simons, M. Circulation 2014, 129, 2557-2566.

5. Kolodgie, F.D.; Gold, H.K.; Burke, A. P.; Fowler, D. R.; Kruth, H. S.; Weber, D. K.; Farb, A.;
Guerrero, L. J.; Hayase, M.; Kutys, R.; Narula, J.; Finn, A. V.; Virmani, R. N. Engl. J. Med.
2003, 349, 2316-2325.
6. Spiegel, S.; Milstien, S. Nat. Rev. Mol. Cell. Biol. 2003, 4, 397-407.
7. Stone, J. R.; Collins, T. Endothelium 2002, 9, 231-238.
8. Ushio-Fukai, M. Cardiovasc. Res. 2006, 71, 226-235.
9. Arai, H. Subcell. Biochem. 2014, 77, 103-114.
10. Pirillo, A.; Norata, G. D.; Catapano, A. L. Mediators Inflamm. 2013, 2013, 152786.
11. Tsimikas, S.; Miller, Y. I. Curr. Pharm. Des. 2011, 17, 27-37.
12. Levitan, I.; Volkov, S.; Subbaiah, P. V. Antioxid. Redox Signal. 2010, 13, 39-75.
13. Dandapat, A.; Hu, C.; Sun, L.; Mehta, J. L. Arterioscler. Thromb. Vasc. Biol. 2007, 27, 2435-
2442.
14. Camaré, C.; Trayssac, M.; Garmy-Susini, B.; Mucher, E.; Sabbadini, R.; Salvayre, R.; Nègre-
Salvayre, A. Br. J. Pharmacol. 2015, 172, 106-118.
15. Harkewicz, R.; Hartvigsen, K.; Almazan, F.; Dennis, E. A.; Witztum, J. L.; Miller, Y. I. J.
Biol. Chem. 2008, 283, 10241-10251.
16. Bochkov, V. N.; Philippova, M.; Oskolkova, O.; Kadl, A.; Furnkranz, A.; Karabeg, E. et al.
Circ. Res. 2006, 99, 900-908.
17. Camaré, C.; Augé, N.; Pucelle, M.; Saint-Lebes, B.; Grazide, M. H.; Nègre-Salvayre, A.;
Salvayre, R. Free Radic. Biol. Med. 2016, 93, 204-216.
18. Heinecke, J. W. Atherosclerosis 1998, 141, 1-15.
19. Stoclet, J. C.; Chataigneau, T.; Ndiaye, M.; Oak, M. H.; El Bedoui, J.; Chataigneau, M.;
Schini-Kerth, V. B. Eur. J. Pharmacol. 2004, 500, 299-313.

20. Nègre-Salvayre, A.; Coatrieux, C.; Ingueneau, C.; Salvayre, R. Br. J. Pharmacol. 2008, 153,
6-20.
21. Aviram, M. Free Radic. Res. 2000, Suppl:33, S85-97.
22. Münzel, T. L.; Gori,T.; Bruno, R. M.; Taddei, S. Eur. Heart J. 2010, 31, 2741-2748.
23. Brigelius-Flohe, R.; Kluth, D.; Banning, A. Mol. Nutr. Food Res. 2005, 49, 1083-1089.
24. Mitra, S.; Deshmukh, A.; Sachdeva, R.; Lu, J.; Mehta, J. L. Am. J. Med. Sci. 2011, 342, 135-
142.
25. Galvani, S.; Coatrieux, C.; Elbaz, M.; Grazide, M. H.; Thiers, J. C.; Parini, A.; Uchida, K.;
Kamar, N.; Rostaing, L.; Baltas, M.; Salvayre, R.; Nègre-Salvayre, A. Free Radic. Biol. Med.
2008, 45,1457-1467.
26. Zhu, Q.; Sun, Z.; Jiang, Y.; Chen, F.; Wang, M. Mol. Nutr. Food Res. 2011, 55, 1375-1390.
27. Bouguerne, B.; Belkheiri, N.; Bedos-Belval, F.; Vindis, C.; Uchida, K.; Duran, H.; Grazide,
M. H.; Baltas, M.; Salvayre, R.; Nègre-Salvayre, A. Antioxid. Redox Signal. 2011, 14, 2093-
2106.
28. Verma, G.; Marella, A.; Shaquiquzzaman, M.; Akhtar, M.; Ali, M. R.; Alam, M. M. J.
Pharm. Bioall. Sci. 2014, 6, 69-80.
29. Delomenède, M.; Bedos-Belval, F.; Duran, H.; Vindis, C.; Baltas, M.; Nègre-Salvayre, A. J.
Med. Chem. 2008, 51, 3171-3181.
30. Vanucci-Bacqué, C.; Chaabouni, S.; Fabing, I.; Bedos-Belval, F.; Baltas, M. Tetrahedron
Lett. 2014, 55, 528-530.
31. a) Brand-Williams, W.; Cuvelier, M. E.; Berset, C. LWT-Food Sci. Technol. 1995, 28, 25-30.
b) Molyneux, P.; Songklanakarin, J. J. Sci. Technol. 2004, 26, 211-219.

32. Sanna, D.; Delogu, G.; Mulas, M. ; Schirra, M.; Fadda, A. Food Anal. Methods 2012, 5, 759-
766.
33. We have checked that comparable IC50 values for Trolox were obtained by UV-absorption or
EPR detections (see Table 1).
34. Hernandez-Vazquez, E.; Castaneda-Arriaga, R.; Ramírez-Espinosa, J. J.; Medina-Campos,
O. N.; Hernandez-Luis, F.; Pedraza Chaverri , J.; Estrada-Soto S. Eur. J. Med. Chem. 2015, 100,
106-118.
35. Lu, J.; Mitra, S.; Wang, X.; Khaidakov, M.; Mehta, J. L. Antioxid. Redox Signal. 2011, 15,
36. Parthasarathy, S.; Raghavamenon, A.; Garelnabi, M. O.; Santanam, N. Methods Mol. Biol.
2010, 610, 403–417.
37. Yagi, K. Chem. Phys. Lipids 1987, 45, 337-351.
38. Salvayre, R.; Auge, N.; Benoist, H.; Negre-Salvayre, A. Biochim. Biophys. Acta 2002, 1585,
213-221.
39. Napoli, C. Ann. N. Y. Acad. Sci. 2003, 1010, 698-709.
40. Vindis, C.; Elbaz, M.; Escarqueil-Blanc, I.; Augé, N.; Heniquez, A.; Thiers, J. C.; Nègre-
Salvayre, A.; Salvayre, R. Arterioscler. Thromb. Vasc. Biol. 2005, 25, 639-45.
41. Cominacini, L.; Pasini, A. F.; Garbin, U.; Davoli, A.; Tosetti, M. L.; Campagnola, M.;
Rigoni, A.; Pastorino, A. M.; Lo Cascio, V.; Sawamura, T. J. Biol. Chem. 2000, 275, 12633-
12638.
42. Costero, A. M.; Gil, S.; Parra, M. P.; Mancini, M. E.; Kneeteman, M. N.; Quindt, M. I.
Tetrahedron Lett. 2015, 56, 3988-3991.

Figure Captions
Figure 1. Synthetic strategy scope - Chemical structure of compound 1a and dialdehyde
precursors 1 and 2.
Figure 2. Capillary tube formation by HMEC-1 stimulated by low oxLDL concentration (20 µg
mL-1) w/wo diaryl phenol derivatives and Trolox (10 µM each). A) Quantification of formed
capillary tubes by HMEC-1 grown on Matrigel. Tube formation was expressed as % of linked
cells to the total cell number. Mean
SEM of
4
separate experiments,
#(oxLDL/control),*(agent/oxLDL) p<0.05. FBS, fetal bovine serum. B) Representative pictures
of tube formation on Matrigel in the presence of the different agents.
Figure 3. Effect of phenolic diaryl derivatives on the intracellular ROS increase evoked by
oxLDL. ROS were determined in HMEC-1 preloaded with the fluorescent H2DCFDA-AM
probe (5 µM), after 40 min incubation with oxLDL (20 µg mL-1), w/wo phenolic diaryl
derivatives or Trolox (10 µM each). Mean
#(oxLDL/control),*(agent/oxLDL) p<0.05.
SEM of 4 separate experiments,
Scheme captions
Scheme 1. Synthesis of the diaryl ether core. Reagents and conditions: a) Cu(0), DMF, reflux; b)
AlCl3, CH2Cl2, rt.

Scheme 2. Synthesis of bis-hydrazones. Reagents and conditions: a) 1-hydrazinylphthalazine
hydrochloride (2 equiv.), EtOH, reflux; b) isonicotinic hydrazide (2 equiv.), EtOH, reflux; c) 3-
methyl-1H-pyrazole-5-carbohydrazide (2 equiv.), EtOH, reflux; d) 2-hydrazinylbenzo[d]thiazole
(2 equiv.), EtOH, reflux; e) benzyl hydrazinecarboxylate (2 equiv.), EtOH, reflux.
Table captions
Table 1. Free radical scavenging activity (DPPH), inhibition (TBARS) of cell-induced LDL
oxidation and cytoprotective effect (MTT) against oxLDL toxicity of the diaryl derivatives at 1
µM and 10 µM.