Substituted pyridine-quinoline ligands as building blocks for neutral rhodium(III) complexes. Synthesis, structural characterization studies and anti-platelet activity towards the Platelet-Activating Factor (PAF)

Article abstract of DOI:10.1016/j.poly.2019.114336

The Friedl?nder condensation reaction was employed to synthesize the new bidentate ligands namely 6-bromo-4-phenyl-2-pyridin-2-yl-quinoline (1) and 4-(4-phenyl-2-(pyridin-2-yl)quinolin-6-yl)phenol (2). These compounds were fully characterized including the X-ray structures of 1 and the protonated form of 2, i.e. 2·HCl·H₂O. We also report the synthesis and spectroscopic characterization of two rhodium(III) complexes of the general formula [Rh(1)Cl₃(CH₃OH)] (3) and [Rh(2)Cl₃(CH₃OH)] (4). The molecular structures of 3·2CH₃OH and 4 were determined by single-crystal X-ray diffraction studies, revealing that these complexes adopt the mer-configuration. The solution stability of 3, 4 was studied by a combination of UV–Vis and ¹H NMR spectroscopic techniques. All compounds were biologically evaluated for their anti- platelet activity using their potency to inhibit the action of PAF (Platelet-Activating Factor), which is an approach of continuous interest in the field. Complexes 3 and 4 were found to be potent PAF inhibitors with IC₅₀ values in the micromolar range (1.0 μM and 3.9 μM respectively). Since PAF is the most potent inflammatory lipid mediator, the PAF-inhibitors and consequently the title compounds may be considered as potent examples in the search for novel anti-inflammatory drugs.

Full text of DOI:10.1016/j.poly.2019.114336

Journal Pre-proofs
Substituted pyridine-quinoline ligands as building blocks for neutral rhodi-
um(III) complexes. Synthesis, structural characterization studies and anti-pla-
telet activity towards the Platelet-Activating Factor (PAF)
Antigoni Margariti, Vasiliki D. Papakonstantinou, George M. Stamatakis,
Constantinos A. Demopoulos, Gregor Schnakenburg, Aikaterini K.
Andreopoulou, Panagiotis Giannopoulos, Joannis K. Kallitsis, Athanassios I.
Philippopoulos
PII:
S0277-5387(19)30781-8
DOI:
https://doi.org/10.1016/j.poly.2019.114336
POLY 114336
Reference:
To appear in:
Polyhedron
Received Date:
Revised Date:
Accepted Date:
5 December 2019
23 December 2019
27 December 2019
Please cite this article as: A. Margariti, V.D. Papakonstantinou, G.M. Stamatakis, C.A. Demopoulos, G.
Schnakenburg, A.K. Andreopoulou, P. Giannopoulos, J.K. Kallitsis, A.I. Philippopoulos, Substituted pyridine-
quinoline ligands as building blocks for neutral rhodium(III) complexes. Synthesis, structural characterization
studies and anti-platelet activity towards the Platelet-Activating Factor (PAF), Polyhedron (2019), doi: https://
doi.org/10.1016/j.poly.2019.114336
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. 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.
©
2019 Elsevier Ltd. All rights reserved.

Substituted pyridine-quinoline ligands as building blocks for neutral rhodium(III)
complexes. Synthesis, structural characterization studies and anti-platelet activity
towards the Platelet-Activating Factor (PAF)
a
b
b
Antigoni Margariti, Vasiliki D. Papakonstantinou, George M. Stamatakis,
Constantinos A. Demopoulos,b Gregor Schnakenburg,c Aikaterini K.
d
d
d
Andreopoulou, Panagiotis Giannopoulos, Joannis K. Kallitsis, and Athanassios
I. Philippopoulos,*a
a
Laboratory of Inorganic Chemistry, Department of Chemistry, National and
Kapodistrian University of Athens, Panepistimiopolis Zografou 15771, Athens,
Greece
b
Laboratory of Biochemistry, Department of Chemistry, National and Kapodistrian
University of Athens, Athens 15771, Greece
c
Institut für Anorganische Chemie, Rheinische Friedrich-Wilhelms-Universität
Bonn, Gerhard-Domagk-Straße 1, D-53121 Bonn, Germany
d
Department of Chemistry, University of Patras, 26504, Patras, Greece
*
Corresponding Author: Athanassios I. Philippopoulos Tel: +30-210-7274697, Fax:
+
30-210-7274782, Email: atphilip@chem.uoa.gr
1

Abstract
The Friedländer condensation reaction was employed to synthesize the new bidentate
ligands namely 6-bromo-4-phenyl-2-pyridin-2-yl-quinoline (1) and 4-(4-phenyl-2-
(pyridin-2-yl)quinolin-6-yl)phenol (2). These compounds were fully characterized
including the X-ray structures of 1 and the protonated form of 2, i.e. 2·HCl·H O. We
2
also report the synthesis and spectroscopic characterization of two rhodium(III)
complexes of the general formula [Rh(1)Cl (CH OH)] (3) and [Rh(2)Cl (CH OH)] (4).
3
3
3
3
The molecular structures of 3·2CH OH and 4 were determined by single-crystal X-ray
3
diffraction studies, revealing that these complexes adopt the mer-configuration. The
1
solution stability of 3, 4 was studied by a combination of UV-Vis and H NMR
spectroscopic techniques. All compounds were biologically evaluated for their anti-
platelet activity using their potency to inhibit the action of PAF (Platelet-Activating
Factor), which is an approach of continuous interest in the field. Complexes 3 and 4
were found to be potent PAF inhibitors with IC values in the micromolar range (1.0
5
0
μM and 3.9 μM respectively). Since PAF is the most potent inflammatory lipid
mediator, the PAF-inhibitors and consequently the title compounds may be considered
as potent examples in the search for novel anti-inflammatory drugs.
1
. Introduction
Platelet-Activating Factor (PAF) is the most potent inflammatory lipid mediator
that participates in a wide range of pathophysiological actions (Scheme 1). Its
systematic name has been identified as 1-O-alkyl-2-acetyl-sn-glycero-3-
phosphocholine [1] however the term PAF includes a broader group of structurally
related (or not related) lipids sharing PAF-like activity [2].
2

Scheme 1. The molecular structure of PAF
The biological activity of PAF is achieved through binding to a G-protein coupled
receptor found on the plasma membrane of cells or intracellular membranes [3]. As a
result, the interaction between PAF and PAFR (PAF-receptor) initiates a cascade of
intracellular signaling pathways, cell/species- dependent, which convert the mediator’s
message to a final cell response [3].
As far as the PAF inhibitors are concerned there is a wide range of compounds that
generally are divided into two categories: the non-specific and the specific PAF
inhibitors [3,4]. The latter (specific PAF inhibitors) constitute a special class of either
naturally occurring compounds such as Gingolides [2,5] or synthetically prepared ones
like rupatadine fumarate [2,6] that antagonize competitively or noncompetitively, the
binding of PAF to its receptor.
Although, organic PAF antagonists exist [3,7], reports on analogous metal-based
inhibitors of PAF are quite rare [3]. This is surprising taking into consideration the
numerous advantages of metal-based therapeutics addressed over the organic analogues
[810]. In addition the role of metal complexes in inorganic medicinal chemistry has
been well documented [1112].
Our interest to investigate the area of metal-based complexes as potential anti-PAF
agents (anti-inflammatory action) stems from the fact that in general metal-based
inhibitors of biological probes, constitute a vast research area owing to the remarkable
characteristics of the metal ions like structural diversity, redox behavior, the possibility
of tuning thermodynamic and kinetic ligand substitution etc. [3,13].
Intrigued by these challenges a systematic approach has been initiated towards the
preparation of potent metal-based inhibitors of PAF, mainly transition metal
complexes. Since PAF is a potent mediator of inflammation, inhibition of its biological
activity (anti-PAF action) by metal-based inhibitors, could contribute to the
3

development of new agents that may have potential anti-inflammatory effects. Notably,
this an approach of continuous interest, as clearly demonstrated from our contributions
in the area, including a series of very effective rhodium [1417], iridium [18] and
ruthenium based inhibitors [1920] of PAF with nanomolar and sub-micromolar
activity. Recent results in this topic have been summarized on a review paper and a
book chapter contribution [3,17] and by other research groups [2122].
The well expressed anti-platelet activity of rhodium(I) and rhodium(III)
complexes bearing quinoxaline and carboxy-pyridine quinoline ligands [15] prompted
us so as to further explore the inhibitory effect of new bifunctional organic ligands
against the PAF-induced aggregation. Substituted pyridine-quinoline ligands,
constitute the basis of this work since (i) their chemistry is closely related to that of the
ligands described in our previous publications [14,15] and (ii) the number of known
metal complexes with this class of organic substances is limited [2325]. In addition,
the knowledge that could be obtained from such study could be evaluated further and
contribute to the structure-activity relationship developed [3,17].
For this purpose, two substituted pyridine-quinoline based ligands, namely the 6-
bromo-4-phenyl-2-pyridin-2-yl-quinoline (1) and the 4-(4-phenyl-2-(pyridin-2-
yl)quinolin-6-yl)phenol (2) were designed and the coordination chemistry of 1, 2 to a
rhodium(III) center was investigated (Scheme 2). Accordingly, the synthesis and
complete characterization of the new versatile chelating ligands 1, 2 and their
rhodium(III) complexes mer-[Rh(1)Cl (CH OH)] (3) and mer-[Rh(2)Cl (CH OH)] (4)
3
3
3
3
is described. Such a synthetic procedure allows us to study the properties of the new
complexes and evaluate the influence of the organic moieties on the biological activity
measured.
In addition, in this manuscript we report for the first time, on the in vitro biological
activity (anti-PAF action) of this category of ligands and of their relevant rhodium(III)
coordination compounds. The inhibitory effect of the new substances was studied on
PAF-induced platelet aggregation toward washed rabbit platelets (WRPs).
4

Scheme 2. Structures of ligands 1 and 2 and atom labeling for NMR assignments.
2
. Materials and methods
2
.1. Reagents and equipments
RhCl ×3H O was purchased from Johnson Matthey Co. and all other reagents and
3
2
solvents from Sigma Aldrich, Alfa Aesar or Acros Organics and were used without
further purification unless otherwise stated. THF was dried over Na/benzophenone and
distilled under argon. (2-Amino-5-bromophenyl)phenyl-methanone [26,27], 4-
(
tetrahydro-2H-pyran-2-yloxy) phenyl boronic acid [28], and Pd(PPh ) [29] were
3 4
synthesized according to the literature reports. The preparation of the organic
precursors 1 and 2 was performed under argon atmosphere, while the synthesis of the
metal complexes 3 and 4 was carried out under aerobic conditions.
Infrared spectra were measured on a Shimadzu IR Affinity-1 spectrometer as
−
1
potassium bromide pellets in the spectral range of 4000–400 cm . The following
abbreviations were used for the intensities of the IR absorption bands: vs = very strong,
s = strong, m = medium, w = weak, br = broad. Elemental analyses were obtained from
the Microanalysis Center of the Institut für Anorganische Chemie Universität Bonn and
from the Instrumental Analysis Laboratory of the University of Patras using a Carlo
1
13
Erba CHNS-O EA1108 Elemental Analyser. H NMR and C NMR spectra were
recorded at 298 K on a Varian 300 MHz and on a Bruker Advance DPX spectrometers
1
13
at 400.13 ( H NMR), and 100.6 MHz ( C NMR) in CDCl (containing TMS as internal
3
standard) and in DMSO-d . J values are given in Hz. The following abbreviations were
6
used for the signal multiplicities: s = singlet, d = doublet, t = triplet, m = multiplet, br
5

=
broad. Absorption spectra were recorded with a CARY 3E UV-Vis spectrometer in
DMSO (DMSO = (CH ) SO) solution. Mass spectra were recorded on a Waters
3
2
Acquity ESI-MS system with a Waters 2966 Photodiode Array detector coupled to a
Waters Micromass ZQ mass spectrometer. A single-quadrupole Quattro micro mass
spectrometer (ACQUITY SQ Detector) equipped with an electrospray ionization (ESI)
interface was used for analytical detection. ESI-MS was operated in positive mode
under the following operating parameters: capillary voltage, 3.5 kV; cone voltage, 30
o
o
V; source temperature, 80 C; desolvation temperature, 300 C; desolvation gas
nitrogen), 5500 L/h; cone gas (nitrogen), 50 L/h. All data were acquired and processed
(
using Masslynx 4.1 software (Waters Corp., MA, Milford, USA). Melting or
decomposition points were determined using an Electrothermal 9100 (IA9000 series)
Digital melting point apparatus and are uncorrected. The samples were sealed in
capillary tubes and heated slowly until the compounds melted or decomposed. Thermal
stability studies were performed with a Thermal analysis Mettler/Toledo TGA/DSC
system under nitrogen atmosphere, between temperature range 25800 °C and with a
temperature increase rate of 10 °C/sec.
2
2
.2. Synthetic procedure and spectral data
.2.1. Synthesis of 6-bromo-4-phenyl-2-pyridin-2-yl-quinoline (1)
Into a 50 mL round bottom flask and under a slow stream of argon, 2-amino-5-
bromophenyl(phenyl)methadone (3 g, 10.86 mmol) and acetylpyridine (2.63 g, 21.72
mmol) were dissolved in a mixture of acetic acid (35 mL) and sulfuric acid 95% (0.1
mL). The reaction mixture was heated under reflux for 3 days. After cooling to room
temperature, the brown solid precipitated was filtered off, washed repeatedly with
methanol and dried under vacuum overnight. Yield: 86% (3.37 g). M.p.: 170.9173.8
°
C. Anal. Calc. for C H N Br: C, 66.50; H, 3.63; N, 7.75. Found: C, 66.47; H, 3.68;
20 13 2
~

1
N, 7.85%. IR (KBr,

in cm ): 3052 (w,  (CHarom)), 1598 (s,  (C=C)), 1547 (m),
1
529 (s), 1487 (s,  (C=N)), 1452 (m), 1227 (w), 1117 (vs,  (CBr)), 1064 (m), 976
s
(w), 883 (m), 833 (m), 800 (m), 780 (m), 755 (m), 700 (s), 619 (s), 600 (w), 576 (w).

1
1
UV-Vis (ε, M cm ): λ (DMSO) = 279 (23529), 314 (12097), 327 (11331), 343
max
1
(
8283). H NMR (CDCl , 400.13 MHz, 298K)  (ppm): 7.38 (td, J = 3.29, 1.59 Hz, 1H,
3
H ), 7.507.57 (m, 5H, H H ), 7.81 (dd, J = 9.0, 2.1 Hz, 1H, H ), 7.90 (td, J = 7.7,
7
9
11
1
6

1
.5 Hz, 1H, H ), 8.09 (d, J = 2.0 Hz, 1H, H ), 8.12 (d, J = 9.0 Hz, 1H, H ), 8.55 (s, 1H,
6 2 3
1
3
1
H ), 8.68 (d, J = 8.0 Hz, 1H, H ), 8.73 (d, J = 4.4 Hz, 1H, H ). C{ H} NMR (CDCl ,
4
5
8
3
1
1
00.6 MHz, 298 K) δ (ppm): 120.02, 121.02, 121.88, 124.26, 127.99, 128.63, 128.72,
1
29.56, 131.91, 132.93, 137.02, 137.71, 147.10, 148.49, 149.23, 155.96. H NMR
(
DMSO-d , 300 MHz, 298K)  (ppm): 7.64 (m, 6H, H +H H ), 7.98 (m, 2H, H +Η ),
6
7
9
11
1
2
8
.10 (td, J = 6.0 Hz, 3Hz, 1H, H ), 8.17 (d, J = 9.0 Hz 1H, H ), 8.52 (s, 1H, H ), 8.68
6 3 4
(
d, J = 6 Hz, 1H, H ), 8.77 (d, J = 6 Hz, 1H, H ). ESI-MS (CH CN): m/z (100%) =
5 8 3
+
3
63.32 [M+H] (calc. 363.05).
2
.2.2. Synthesis of 4-(4-phenyl-2-(pyridin-2-yl)quinolin-6-yl)phenol (2)
g (8.31 mmol) of ligand 1 and 4-(tetrahydro-2H-pyran-2-yloxy) phenyl boronic
3
acid (2.21 g, 9.97 mmol) were dissolved in 70 mL of degassed toluene under inert
atmosphere, followed by the addition of K CO (6.88 g, 49.86 mmol) and Pd(PPh )
3 4
2
3
(0.24g, 0.21 mmol) and the resulting mixture was refluxed for 3 days. The solution was
then cooled to ambient temperature, filtered from filter paper and extracted with ethyl
acetate and deionized water. The organic layer was dried over MgSO , filtered off and
4
evaporated under reduced pressure before drying in a vacuum oven overnight affording
1
3
.23 g of the intermediate 2-THP. Yield: (85%). H NMR (CDCl , 400.13 MHz, 298K)
3
δ (ppm): 1.661.75 (m, 3H, H +H ), 1.89 (m, 2H, H ), 2.02 (sext, J = 7.0 Hz, 1H,
1
5
16
17
H ), 3.62 (dt, J = 4.3, 3.0 Hz, 1H, H ), 3.92 (td, J = 10.5, 3.1 Hz, 1H, H ), 5.47 (t, J
1
5
18
18
=
3.2 Hz, 1H, H ), 7.13 (d, J = 4.9 Hz, 2H, H ), 7.36 (t, J = 5.8 Hz, 1H, H ), 7.517.58
1
4
12
7
(
m, 5H, H H ), 7.63 (d, J = 6.3Hz, 2H, H ), 7.90 (td, J = 7.7, 1.8 Hz, 1H, H ), 7.98
9
11
13
6
(
dd, J = 8.8, 2.0 Hz, 1H, H ), 8.09 (d, J = 2.0 Hz, 1H, H ), 8.29(d, J = 8.7 Hz, 1H, H ),
1 2 3
8
.52 (s, 1H, H ), 8.70 (d, J = 8.7 Hz, 1H, H ), 8.72 (d, J = 4.8 Hz, 1H, H ).
4 5 8
Subsequently, the obtained solid 2-THP was dissolved in THF (150 mL) and then
concentrated HCl (2 mL) was added. The color of the mixture was immediately turned
to an intense red and it was stirred at room temperature overnight under argon. After
that period a yellow solid had precipitated. Solvents were removed under reduced
pressure and the remaining solid was stirred in deionized water, filtered off, washed
with deionized water and then the solid was dried under vacuum overnight. Τhe
obtained solid was redissolved in THF and was treated with 20 mL of an 0.5M aqueous
solution of Na CO . The mixture was stirred at ambient temperature for 6 h and then
2
3
7

the organic solvent was removed under pressure. The obtained solid was washed
extensively with deionized water, filtered off and dried under vacuum overnight. Yield:
o
9
7
6% (3.10 g). M.p.: 213.5217.1 C. Anal. Calc. for (2) × 1.0(H O), C H N O : C,
2
26 20
2
2
~

1
9.58; H, 5.14; N, 7.14. Found: C, 79.64; H, 4.55; N, 7.04%. IR (KBr,

in cm ): 3423
(
br,  (ΟΗ)), 3056 (w,  (CHarom)), 2925 (w,  (CHaliph)), 1610 (vs), 1586 (vs), 1521
s), 1490 (vs,  (C=N)), 1438 (m), 1363 (m), 1273 (vs), 1260 (w,  (OH)), 1281 (s),
(
1
176 (m,  (CO)), 1027 (w), 895 (w), 825 (vs), 797 (s), 760 (s), 702 (s), 620 (s), 590

1
1
(
m), 543 (m), 529 (m), 440 (w). UV-Vis (ε, M cm ):  (DMSO) = 289 (34160),
max
1
3
1
2
8

7
7
43 (14460), 358 (14833). H NMR (DMSO-d , 300 MHz, 298K)  (ppm): 6.89 (d, J =
6
0.0 Hz, 2H, H ), 7.507.68 (m, 6H, H +H +H H ), 8.00 (s, 1H, H ), 8.048.11 (m,
1
2
7
13
9
11
2
H, H +H ), 8.24 (d, J = 8.0 Hz, 1H, H ), 8.48 (s, 1H, H ), 8.67 (d, J = 6.0 Hz, 1H, H ),
6
1
3
4
5
1
.73 (d, J = 6.0 Hz, 1H, H ), 9.69 (s, 1H, OH). H NMR (CDCl , 400.13 MHz, 298K)
8
3
(ppm): 5.04 (s, 1H, OH,), 6.91 (d, J = 8.4 Hz, 2H, H ), 7.37 (t, J = 4.0 Hz, 1H, H ),
1
2
7
.517.57 (m, 5H, H H ), 7.63 (d, J = 7.6 Hz, 2H, H ), 7.90 (t, J = 7.6 Hz, 1H, H ),
9
11
13
6
.97 (d, J = 8.8 Hz, 1H, H ), 8.07 (s, 1H, H ), 8.29 (d, J = 8.8 Hz, 1H, H ), 8.52 (s, 1H,
1
2
3
1
3
1
H ), 8.70 (d, J = 7.6 Hz, 1H, H ), 8.73 (d, J = 4.8 Hz, 1H, H ). C{ H} NMR (CDCl
3,
4
5
8
1
1
1
00.6 MHz, 298K)  (ppm): 115.87, 119.71, 122.00, 122.49, 124.00, 126.60, 126.89,
28.27, 128.48, 128.56, 128.97, 129.58, 130.31, 131.98, 132.08, 132.60, 137.10,
38.25, 139.19, 147.48, 149.04, 149.31, 155.06, 156.00, 156.29. ESI-MS (CH CN) :
3
+
+
m/z (100%) = 375.48 [M+H] (calc. 374.14), 397.46 [M+Na] (calc. 397.13).
.2.3. Synthesis of mer-[Rh(1)Cl (CH OH)] (3)
2
3
3
Into a 50 mL round bottom flask, 56 mg (0.21 mmol) of RhCl ×3H O was
3
2
dissolved in 13 mL of warm CH OH giving a brown-red solution. An equimolar amount
3
(77 mg) of 1 was dissolved in 13 mL of warm methanol and subsequently was added
slowly and under stirring to the homogenous solution above. During the addition, the
color of the solution changed to bright orange and the solution was heated to a gentle
reflux for about 1 h. Upon cooling at ambient temperature a yellow-orange
microcrystalline solid was precipitated and was dried in vacuo affording 71 mg of
analytically pure complex 3 (yield: 55.4%). The clear orange filtrate was left to stand
in a closed vial at ambient temperature, affording needle-like orange crystals suitable
for single crystal X-ray diffraction study after a period of two weeks. The crystals were
8

collected from the mother liquor and were dried in vacuo affording a further crop of the
desired product (27 mg). Total yield: Yield: 77% (98 mg). Anal. Calc. for (3) ×
1
.0(H O), C H BrCl N O Rh: C, 40.62; H, 3.09; N, 4.50. Found: C, 40.34; H, 3.00;
2 21 19 3 2 2
~

1
N, 4.70%. IR (KBr,
437 (m), 1383 (s), 1361 (w), 1261 (w), 1221 (w), 1165 (m), 1130 (w), 1075 (w), 985
w), 886 (m), 827 (w), 785 (s), 756 (m), 698 (m), 646 (w), 601 (w), 544 (w). UV-Vis

in cm ): 3065 (w,  (CHarom)), 1591 (m), 1540 (m), 1481 (vs),
1
(
(

1
1
1
ε, M cm ):  (DMSO) = 297 (28829), 342 (15196), 355 (19291), 398 (3367). H
max
NMR (DMSO-d , 300 MHz, 298K)  (ppm): 3.17 (s, 3H, CH ), 4.08 (br.s, 1Η, OH),
6
3
7
.697.77 (m, 5H, H H ), 7.88 (t, J = 6 Hz, 1H, H ), 8.03 (s, 1H, H ), 8.24 (d, J = 6
9
11
7
2
Hz, 1H, H ), 8.32 (t, J = 6 Hz, 1H, H ), 8.75 (s, 1H, H ), 8.96 (d, J = 6 Hz, 1H, H ),
1
6
4
3
9
.62 (d, J = 9 Hz, 1H, H ), 9.81 (d, J = 6 Hz, 1H, H ).
5
8
2
.2.4. Synthesis of mer-[Rh(2)Cl (CH OH)] (4)
3
3
In a 50 mL round bottom flask, 54 mg (0.205 mmol) of RhCl ×3H O was dissolved
3
2
in 5 mL of warm CH OH. An equimolar amount (77 mg) of 2 dissolved in warm
3
CH OH (10 mL) was added under stirring to the homogenous red-brown solution and
3
the mixture was then heated under reflux for about 1 h affording an orange solution.
After cooling at room temperature, the clear solution was left for slow evaporation at
room temperature. Two days later, orange needle-like crystals were formed and some
of them were suitable for single-crystal X-ray diffraction study. The rest of the
crystalline material was collected by filtration and dried in vacuo affording 30 mg
(yield: 40%) of 4. A further amount of the compound was collected upon slow
evaporation of the mother liquor at ambient temperature. Total yield: 60% (45 mg).
Anal. Calc. for Calc. for for (1) × 3(H O), C H Cl N O Rh: C, 48.40; H, 4.22; N,
2
27 28
3
2
5
~

1
4
.18%. Found: C, 48.68; H, 3.87; N, 4.37%. Calc. IR (KBr,

in cm ): 3334 (br, 
(ΟΗ)), 1605 (s), 1588 (s), 1520 (m), 1488 (vs), 1443 (w), 1385 (m), 1359 (m), 1259 (w,

(OH)), 1180 (m,  (CO)), 1131 (w), 1110 (w), 894 (w), 873 (w), 826 (s), 782 (s),

1
1
7
60 (s), 704 (s), 629 (w), 604 (w), 529 (w). UV-Vis (ε, M cm ):  (DMSO) = 312
max
1
(
47186), 367 (13246), 392 (18466), 426 (12073). H NMR (DMSO-d , 300 MHz, 298K)
6

(ppm): 3.17 (s, 3H, CH ), 4.08 (br, 1H, OH), 6.92 (d, J = 9 Hz, 2H, H ), 7.59 (d, J
3
12
=
6 Hz, 2H, H ), 7.697.83 (m, 6H, H + H H ), 8.01 (s, 1H, H ), 8.278.37 (m, 2H,
1
3
7
9
11
2
9

H Η ), 8.67 (s, 1H, H ), 8.93 (d, J = 9 Hz, 1H, H ), 9.70 (d, J = 9 Hz, 1H, H ), 9.80
1
+
6
4
3
5
(
br.s, 1H, H ), 9.83 (br.s, 1H OH of PhOH).
8
2
.3. X-ray analysis
The data collections were performed on a Bruker D8-Venture (3), a Nonius Kappa
CCD (2, 4), and a Bruker X8-KappaApexII diffractometer by using Cu-Kα- (1, λ =
.54178 Å) or Mo-Kα radiation (λ = 0.7103 Å), generated by a sealed tube. The
1
diffractometers were equipped with a low-temperature device; the measurements where
performed at 123(2) K).
Intensities were measured by fine-slicing  and -scans and corrected for
background, polarization, absorption and Lorentz effects [30].
The structures were solved by intrinsic phasing methods implemented in
Sheldrick’s XT program and refined anisotropically by the least-square procedure
implemented in the SHELX program system [31,32]. Hydrogen atoms were included
using the riding model on the bound carbon atoms.
Crystallographic data and refinement conditions for compounds 1–4 are
summarized in Table 1.
Table 1
Crystal and refinement data for 1, 2∙HCl∙H O, 3∙2CH OH and 4.
2
3
1
2∙HCl∙H O
3∙2CH OH
4
2
3
Empirical formula
Molecular weight
Crystal color
C H BrN
361.23
colourless
C H ClN O
428.90
yellow
C H BrCl N O Rh C H Cl N O Rh
23 25 3 2 3 27 22 3 2 2
2
0
13
2
26 21
2
2
666.62
615.72
orange
orange
Crystal size (mm3)
0.30×0.08× 0.02 0.24 × 0.12×0.11 0.24×0.10×0.06
123(2)
0.34× 0.10×0.07
123(2)
T (K)
123(2)
123(2)
Crystal system
Space group
orthorombic
Pbca
orthorombic
Pbca
monoclinic
trigonal
P2 /c
P3 21
1
2
Unit cell dimensions



(Å)
(Å)
(Å)
8.0304(3)
19.3245(7)
19.4156(7)
17.3460(3)
10.1900(2)
23.7866(5)
14.1125(4)
13.9370(3)
27.7042(8)
28.2138(12)
28.2138(12)
8.4237(4)
1
0







90
90
90
90
90
90
111.344(2)
90
90
90
90
120
3
V (Å )
3018.57(19)
8
4204.42(14)
8
5075.3(2)
8
5807.1(6)
6


3
calc (g cm )
1.590
1.355
1.745
1.056

1

(mm )
F(000)
θ range (
3.698(Cu-K)
1456.0
0.208(Mo-K)
1792
2.592(Mo-K)
2656
5.468 to 52
23710
9729
6/611
0.667(Mo-K)
1860
3.334 to 58.408
59766
9983
88/376
2
9.09 to 34.96
11737
2626
0/208
0.0616
2.35 to 28.00
59408
5077
0/292
0.0406
Reflections collected
Unique reflections
Parameters refined
R1 [2I)]
0.0425
0.0465
wR (all data)
0.1748
1.029
0.1250
1.062
0.0943
1.089
0.1229
1.050
2
Goodness-of-fit on F2
Largest diff. peak/hole 0.76/2.01
0.624/ –0.369
0.59/0.60
1.28/–0.61

3
(e Å )
2
.4. Biological experiments
The biological experiments PAF- induced platelet aggregation, and the inhibitory
effect of the synthetic compounds) were performed according to well described
protocols [1,14]. Briefly, for the washed rabbit platelets (WRPs) preparation, rabbit
platelets (platelet rich plasma), were washed on Ficoll-Paque cushions and adjusted to
2
.5 x 10-8 platelets/ml of Tyrode’s buffer, pH 7.2. The dilution of PAF was prepared
in pyrogen-free 0.15 M NaCl containing 2.5 mg/ml of crystallized bovine serum
albumin, required for dispersion of PAF. The compounds under investigation were first
dissolved in DMSO and aliquots of these solutions were then added in solutions (2.5

mg BSA/mL) of saline. Microliters of the dilution of PAF (final concentration, 10 10
M, in the cuvette) were added in the cuvette of the aggregometer and calculate the
height of the platelet aggregation. The inhibitory effect of the rhodium precursor
RhCl ×3H O, the ligands and the rhodium(III) coordination compounds prepared,
3
2
towards the washed rabbit platelets (WRPs) aggregation experiments, was carried out
by literature methods [15]. The aggregometer was from Chrono-Log Corporation,
(Haverton, PA). Bovine serum albumin (BSA), PAF and analytical solvents for the
biological assays were purchased from Sigma. The measurements were repeated at least
twice for each sample concentration in three different WRPs preparations, in order to
1
1

ensure reproducibility. All experiments were also followed by appropriate control tests
of the solvent used (saline solution of BSA, DMSO aliquots in a saline solution of BSA-
2
.5 mg BSA/mL saline- or just in saline), in WRPs.
3
. Results and Discussion
3
.1. Synthesis and characterization of the organic ligands 12
Compound 1 was prepared in 86% yield following a Friedländer condensation
reaction of 2-amino-5-bromobenzophenone (otherwise known as (2-amino-5-
bromophenyl)phenyl-methanone)) and 2-acetylpyridine under acidic condition. The
synthetic procedure is outlined in Scheme 3. Although this ligand has been employed
in previous studies for the preparation of 6-(4-vinylphenyl)-2-(2-pyridine)-4-
phenylquinoline employed as a monomer in radical polymerizations, as well as for the
synthesis of 6-(9-carbazolyl)-4-phenyl-2-(2-pyridyl)-quinoline [25], its synthetic
protocol and spectroscopic characterization were not reported up to now [33,34].
A complete spectroscopic characterization of 1 is now included supported by mass
spectrometric methods. The electrospray mass spectrum of 1 exhibits a base peak at m/z
+
3
63.32 for the [M+H] ion with an isotope pattern characteristic of the presence of one
bromine atom and is depicted in the supporting information (Fig. S1).
1
13
The H and C NMR spectra of 1 in CDCl display the number of expected
3
resonances (Figs. S2S3). The characteristic down field doublet resonances for the H8
and H protons of the pyridine ring are observed at  8.77 and  8.68 respectively, while
5
the singlet resonance at  8.52 is due to the H proton of the benzene ring of 6-Br-
4
quinoline.
In the FTIR spectrum quite informative is the very strong stretching vibration for


the  (CBr) symmetric mode at 1116 cm 
Scheme 3. Synthesis of 6-bromo-2-(2-pyridine)-4-phenylquinoline (1).
1
2

Suitable colorless single crystals of 1 were obtained upon slow evaporation of a
DMSO-d solution of the compound at ambient temperature. The molecular structure
6
of 1 is shown in Fig. 1.
Fig. 1. Τhe molecular structure of 1. Hydrogen atoms are omitted for clarity. The
ellipsoids were plotted at 50% probability. Selected bond lengths (Å) and angles (°):
BrC5 = 1.906(4), N1C2 = 1.365(5), N1C9 = 1.324(5), C9C16 = 1.488(5), N2C16
=
1.353(5), N2C17 = 1.341(5), N1C9C16 = 117.2(4), C8C7C10 = 120.2(4).
Compound 1 crystallizes in the orthorhombic Pbca space group and as expected, the
two nitrogen atoms of the bipyridine group of the ligand adopt a trans-configuration.
The pyridine ring close to the quinoline ring deviates from planarity (angle between the
least square planes of rings containing atoms N1 and N2 is 19.62°). The phenyl group
connected to C7 is also deviating from planarity (angle between the least square planes
of rings containing atoms N1 and C10 is 57.62°) and is further involved in typical
intramolecular CH∙∙∙ contacts [35] with the phenyl ring (C10C15) from an adjacent
molecule (distance of (C11H11∙∙∙centroid) = 2.917 Å). In the crystal packing of 1 the
bromine atom displays weak CH∙∙∙Br contacts (C8H8∙∙∙Br = 3.889 Å, H8∙∙∙Br = 3.262
Å; C18H18∙∙∙Br = 3.875 Å, H18∙∙∙Br = 3.244 Å) that is in accord with the literature
[36]. Finally the crystal is stabilized by rather weak  stacking interactions between
the phenyl and pyridine ring (distance between centroids (C1C6) and (N1C1C9) of
3
.962 Å) as shown in the supporting information (Fig. S4) [37].
The analytically pure compound 1 obtained through the route described previously,
was further used for the synthesis of the hydroxylphenyl functionalized ligand 2, in a
two-step reaction as presented in Scheme 4. The first step leads to the isolation of the
1
3

tetrahydropyran (THP)-protected intermediate 2-THP, through the Suzuki coupling of
with the 4-(tetrahydro-2H-pyran-2-yloxy) phenyl boronic acid using Pd(PPh ) as
1
3
4
1
catalyst. The purity of this intermediate was checked by H NMR in CDCl (Fig. S5a)
3
proving the successful formation of the 2-THP derivative which was used without any
further purification. Subsequent removal of the tetrahydropyranyl protective group
under acidic conditions, led to the desired ligand 2 in 96% yield (Fig. S5b).
Scheme 4. Synthesis of 6-(4-yl-phenol)-2-(2-pyridine)-4-phenylquinoline (2).
The molecular composition of 2 was unambiguously confirmed by electrospray
mass spectrometry (Fig. S6) exhibiting peaks at m/z 375.48 and 397.46 corresponding
+
+
1
13
to [M+H] (basic peak) and [M+Na] . The H and C NMR spectra of 2 in DMSO-d ,
6
are presented in the supporting information (Figs. S7S8). The low field single
resonance of the phenolic hydroxyl proton is oserved at  9.69. An important feature in
1
the H NMR spectrum of 2 is the presence of the single resonance at  8.48 that is
characteristic of the H proton of the ligand. The doublet resonances at  6.89 and 
4
7
.52 are typical for the H and H protons of the PhOH group. The latter (doublet
1
2
13
resonance for H ) is overlapped with the resonance signal of H and the protons of the
1
3
7
phenyl group (H H ) respectively. All other resonances appear in the expected
9
11
1
1
region; their assignment was based on a H H COSY NMR spectrum, confirming
undoubtedly the molecular structure of 2.
Despite repeated crystal growth attempts with this compound only microcrystals
or powders were obtained. Therefore we tried with the protonated form of the ligand
(
2·HCl), that is the compound obtained prior to treatment with Na CO3(aq) (see
2
1
experimental). The IR and H NMR spectra of 2·HCl that was used for crystallization,
1
4

are included in the supporting information (Figs. S9S10). Finally, yellow single
crystals of 2·HCl·H O suitable for single-crystal X-ray diffraction studies, were
2
obtained upon slow diffusion of diisopropyl ether to a methanol solution of 2·HCl at
ambient temperature.
Compound 2·HCl·H O is crystallizing in the orthorombic crystallographic system,
2
Pbca space group and the molecular structure is depicted in Fig. 2.
Fig. 2. The molecular structure of 2∙HCl∙H O. Cl counter anion and the crystalline
2
water molecule are omitted for clarity. The ellipsoids were plotted at 50% probability.
Selected bond lengths (Å) and angles (°): O1H14 = 0.97(2), O1C24 = 1.3621(18),
N1C1 = 1.3560(19), N1H1A = 0.941(18), N2C6 = 1.3306(18), N2C10 =
1
1
.3558(19), C5N1C1 = 123.00(13), C5N1H1A =120.4(11), N2C6C1 =
15.45(12), N2C10C20 = 118.03(4).
The phenyl substituent on C8 is twisted out of the plane of the bipyridine (bpy)
moiety (angles between the least square planes of rings containing N1, N2 and
C11C16 are 49.77°). Also the phenol group on C18 deviates from planarity (angles
between the least square planes of rings containing C18, C21 are 36.0°). In the
bipyridine moiety the two nitrogen atoms about the biaryl bond are in the cis-position
and this preference could be attributed to the intramolecular hydrogen bonding of N2
to the adjacent H1A of the N1 atom (N1H1A∙∙∙N2 = 2.217 Å, bond angle = 106.76°)
[
38].
The packing in the crystal is such that four sets of ribbons (eight molecules) are
stabilized by weak intermolecular  stacking (distance between centroids are in the
range of 3.927 to 4.024 Å). Each set of ribbons is almost orthogonal to one another.
The phenolic oxygen atom of the ligand displays non classical intermolecular hydrogen
1
5

bonding interactions with an adjacent molecule, (distance of O1∙∙∙H5AC5 = 2.497 Å
N1 ring); bond angle of 125.63°) [39] and characteristic O1H1∙∙∙ weak hydrogen
(
bonding interactions at a distance of 3.402 Å (O1H1∙∙∙ centroid (ring Ν1)) that overall
stabilize the structure in the crystal [40]. The network that is formed is reinforced by
intermolecular hydrogen bonding interactions between the oxygen atoms of the phenol
group with solvated water molecules (O2water) in the lattice (O1H1∙∙∙O2water = 2.699Å)
and the Cl counter ions (O2H2B∙∙∙Cl = 2.433 Å) respectively. Overall these
interactions are shown in the supporting information (Fig. S11).
Although 2·HCl·H O is the structure of the protonated form of the ligand precursor
2
2, the information collected upon structure elucidation is of importance in relation to
the nature of the ligand. In addition, a Cambridge Crystallographic Database search
revealed that its crystal structure has not been reported previously, while
crystallographic data on this class of compounds are quite rare. So far, only the
molecular structures of 2-(2-Pyridyl)-4-phenyl-quinoline and of its relevant complex
with Zn(II) have been reported [25].
Finally it should be mentioned, that 1, 2 are easily prepared materials, in high yields
and in a multi-gram scale which can be stored for weeks to months in common
desiccators. Accordingly they can be potentially used as building blocks for the
synthesis of a number of coordination compounds, as is the case herein for the Rh(III)
coordination compounds 3 and 4 respectively.
3
.2. Synthesis and characterization of the rhodium(III) complexes 34
The octahedral rhodium(III) complexes 3 and 4 were prepared according to the
synthetic protocol described previously by our group [15]. The reaction of RhCl ×3H O
3
2
with an equimolar amount of 1 or 2 was taken place in hot methanol (Scheme 5).
Subsequent work up of the reaction mixture resulted to the isolation of analytically pure
mer-[Rh(1)Cl (CH OH)] (3) and mer-[Rh(2)Cl (CH OH)] (4) as yellow-brown (3) and
3
3
3
3
orange (4) microcrystalline solids in 77% and 60% yield respectively. The structures of
and 4, as the mer isomers, was undoubtedly confirmed from a single crystal X-ray
diffraction analysis performed (vide infra). Complex 3 is soluble in methanol, acetone,
3
1
6

CH CN, THF, DMF and DMSO, while 4 dissolves in acetone, methanol, DMF and
3
DMSO. Both compounds are sparingly soluble in water.
Scheme 5. The synthetic procedure of the rhodium(III) complexes 3 and 4.

1
The IR spectrum of 3 in the region of 1600–400 cm exhibits strong bands for the
C=C, C=N stretching vibrations and the pyridyl ring deformation bands of 1. The

1
presence of a medium intensity band at 1130 cm could be presumably attributed to
the  (C–Br) symmetric mode.
1
The H NMR spectrum of 3 recorded in DMSO-d , is consistent with the presence
6
of a single ligand environement (Fig. S12). Assignment of the proton resonances
1
1
became apparent with the help of a H H COSY NMR experiment performed.
Coordination of 1 has been identified unambiguously by the pronounced downfield
shift of the H and H protons at  9.81 and  9.62, in comparison to the corresponding
8
5
resonance signals of the free ligand at  8.77 and  8.68 respectively ( = 1.04 and
1
0
.94 ppm). In addition, the double resonance found for the H protonat  8.96 is shifted
3
towards lower fields by 0.79 ppm as compared to that of 1, while the multiplet
1
A detailed comparison with the NMR data of the free ligand (in the same solvent)
1
cannot be made. Single crystal X-ray diffraction studies and H-NMR studies for 3 have
shown that the free ligand exists in a trans conformation. Upon coordination to a metal
center the cis conformation is adopted.
1
7

resonances observed at δ 8.32 and  7.88 are assigned to the adjacent H and H protons.
6
7
In the aliphatic region, only two resonances were observed at 3.17 (CH ) and  4.088
3
(OH) that can be attributed to free methanol [41].
In the IR spectrum of 4 characteristic is a broad vibration for the  (OH) of the

1
phenol group at 3334 cm that is typical for the intermolecular hydrogen bonded OH

1
stretch. The  (OH) bending vibration probably occurs at 1259 cm and the  (CO)

1
stretching mode is present as a medium intensity band at 1179 cm .
1
The solution H NMR spectrum of 4 (crystals of 4 dissolved in DMSO-d ) in the
6
aromatic region displays sharp, well resolved and intense resonances that are
characteristics for the presence of 2 (Fig. S13). Ligand coordination becomes evident
by the remarkable low frequency shift of the signals due to H ( 9.80) and H ( 9.70).
8
5
In the free ligand, these resonance signals are at  8.73 and  8.67 respectively.
1
1
According to the H H COSY NMR spectrum, the singlet resonance at  9.80 is
overlapped with a rather broad resonance signal at  9.83 that is attributed to the
phenolic hydroxyl proton (OH). In addition, the doublet resonance for H ( 8.93) is
3
downfield shifted when compared to that of compound 2 at  8.24 (Δδ = 0.69 ppm),
denoting that the coordination environment around it has been altered significantly. The
H and H protons appear as a multiplet resonance between 8.278.37 ppm, while the
6
1
H resonance at δ 7.69–7.83 is overlapped with the phenyl group protons H H . In
7
9
11
1
the aliphatic region of the H NMR spectrum of complex 4, only the presence of free
methanol was detected at  3.17 (CH ) and  4.08 (OH), as reported previously for
3
the analogues complex 3.
3
.3. Thermal behavior of complexes 34
Thermal stability of 3, 4 (in the crystalline form) was examined by means of a
o
thermogravimetric analysis (TGA) performed in the temperature range 25800 C
under nitrogen atmosphere (Figs. S14S15). The thermal analysis data of the
o
rhodium(III) complexes indicate that, decomposition starts at 25 C and the first weight
o
o
loss (128.2 C for 3; temperature range 25 C for 4) corresponds to loss of one
coordinated methanol molecule (Obs. 6.0 and 5.5%, Calcd. 5.3 and 5.3 % for 3 and 4
respectively).
1
8

For complex 3 a detailed analysis could not be undertaken since thermal
o
decomposition continues over 800 C. Weight losses cannot be assigned to individual
moieties and thus the information provided is rather limited. On the other hand,
o
complex 4 continues to decompose thermally, up to 800 C through a second plateau
o
o
(
between 128 C–265 C) and intermediate non-well separated steps. Attribution of
weight losses to individual molecule moieties is not possible for any of these steps.
o
o
However, the total weight loss (75.56 %) from 125 C to 800 C could be assigned to
loss of the C H NO Cl moiety, that corresponds to the fragment [(2)Cl ] (Calcd.
2
6
17
2
3
3
o
7
8.03%). At 800 C, the total weight loss of 18.9 % corresponds to Rh O (Calcd. 20.26
2
3
%).
3
.4. Structural characterization of complexes 34
The solid-state structures of the rhodium(III) complexes 3∙2CH OH and 4 were
3
determined by single-crystal X-ray diffraction. Orange needle-like crystals of
·2CH OH were grown upon standing of a methanol solution of the complex in a closed
3
3
vial at ambient temperature over a period of two weeks. This compound is crystallizing
in the monoclinic crystallographic system, P2 /c space group. The molecular structure
1
of 3 is presented in Fig. 3 and selected bond distances and angles are listed in Table 2.
Two independent molecules of 3·2CH OH with slightly different bonding parameters
3
were found in the asymmetric unit. Complex 3 shows a distorted octahedral geometry
with three chlorido ligands in the mer configuration and the ligand 1 coordinated in a
bidentate fashion. The remaining site of the coordination sphere is completed by a
methanol molecule bound to each rhodium(III) center. The Rh(1)–O(1) bond length of
the coordinated methanol molecule is at 2.080(3) Å and the C(21)–O(1)–Rh(1) angle
at 124.1(3)°. Typical bond angles around the metal center (one of the two independent
molecules) manifesting distortion, are these of Cl(3)–Rh(1)–Cl(1) = 89.63(5)°, O(1)–
Rh(1)–Cl(1) = 93.23(11)°, O(1)–Rh(1)–Cl(3) = 84.86(10)° and Cl(2)–Rh(1)–Cl(1) =
1
78.58(4)° respectively. The Rh(1)–N(2) bond distance of 1.987(4) Å is significantly
shorter compared to that of Rh(1)N(1) = 2.084 Å(4), reported for the mer-
RhCl (bpy)(CH OH)]·CH OH complex [41]. The bond lengths of Rh(1)Cl(1) =
[
3
3
3
1
9

2
.338(14) Å, Rh(1)Cl(2) = 2.334(13) Å and Rh(1)Cl(3) = 2.326(13) Å compare well
with the values reported for other rhodium(III) complexes [42].
Fig. 3. Molecular structure of 3 (one of the two molecules in the crystal). Hydrogen
atoms and solvated methanol molecules are omitted for clarity. The ellipsoids were
plotted at 50% probability.
The molecular structure of 3 is stabilized by intermolecular OH···O and
OH···Cl hydrogen bonding interactions. Coordinated and free methanol molecules
participate in these interactions (Fig. S16), while selected hydrogen bonding
interactions are shown in the supporting information (Table S1). For example, the
O(5)H(5)···Cl(4)) hydrogen bonding interaction of 2.525 Å (bond angle (<) OH···Cl
=
147.7°) compares well with the one reported for the fac-trichloridotris(trimethyl-
phosphane-κP)rhodium(III) methanol hemihydrate, (OH···Cl bonding interaction of
.57 Å (bond angle (<) OH···Cl = 157°) [43]. Complex 3 exhibits characteristic
2
CH∙∙∙ interactions (Fig. S17) that overall stabilize the structure in the crystal (distance
of (C21H21C)∙∙∙centroid (ring C(28)) = 3.650 Å and that of (C42H42B)∙∙∙centroid
(ring C(7)) = 3.239 Å).
Table 2
Selected bond lengths (Å) and bond angles (°) for 3∙2CH OH and 4.
3
bond lengths
bond angles
a
3
∙2CH OH
3
Rh(1)–Cl(1) 2.3383(14) Cl(2)–Rh(1)–Cl(1) 178.58(4)
Rh(1)–Cl(2) 2.3336(13) Cl(3)–Rh(1)–Cl(1) 89.63(5)
Rh(1)–Cl(3) 2.3261(13) Cl(3)–Rh(1)–Cl(2) 89.99(5)
Rh(1)–N(1) 2.084(4)
O(1)–Rh(1)–Cl(1) 93.23(11)
2
0

Rh(1)–N(2) 1.987(4)
Rh(1)–O(1) 2.080(3)
C(14)–Br(1) 1.899(4)
O(1)–Rh(1)–Cl(2) 88.09(11)
O(1)–Rh(1)–Cl(3) 84.86(10)
O(1)–Rh(1)–N(1)
99.41(14)
N(1)–C(1)
N(1)–C(6)
O(1)–C(21) 1.433(7)
1.333(6)
1.387(6)
N(1)–Rh(1)–Cl(1) 89.14(11)
N(1)–Rh(1)–Cl(3) 175.62(11)
C(21)–O(1)–Rh(1) 124.10(3)
4
Rh–Cl(1)
2.3387(14) Cl(2)–Rh–Cl(3)
2.3225(14) Cl(2)–Rh–Cl(1)
2.3533(15) Cl(1)–Rh–Cl(3)
178.48(5)
91.10(6)
90.42(6)
87.13(12)
91.64(14)
88.38(14)
177.03(19)
95.26 (12)
89.83(12)
122.2(4)
Rh–Cl(2)
Rh–Cl(3)
Rh–N(1)
Rh–N(2)
Rh–O(2)
N(1)–C(1)
N(1)–C(5)
1.974(4)
2.041(4)
2.106(4)
1.346(6)
1.339(7)
O(2)–Rh–Cl(1)
O(2)–Rh–Cl(2)
O(2)–Rh–Cl(3)
O(2)–Rh–N(1)
N(1)–Rh–Cl(1)
O(2)–C(27) 1.461(10) N(1)–Rh–Cl(3)
O(1)–C(24) 1.364(6) C(27)–O(2)–Rh
a
Bond lengths and angles of one of the two independent molecules.
Finally, the crystal packing in the structure of 3 is such that symmetry related
molecules are separated by typical  stacking interactions at a distance of 3.596 Å
and 3.844 Å respectively (Fig. 4).
Although rhodium(III) complexes with coordinated solvent molecules have been
reported [444546], the number of crystallographically characterized methanol solvate
complexes is quite rare. A crystallographic search in the Cambridge Structural Database
1
1
revealed only the mer-[Rh(L )Cl (CH OH)] (L = 2-(2′-pyridyl)quinoxaline) [16] and
3
3
the mer-[Rh(bpy)Cl (CH OH)]·(CH OH) [41] complexes respectively.
3
3
3
2
1

Fig. 4. Intermolecular  stacking interactions in complex 3. Carbon, gray; hydrogen,
light gray; nitrogen, blue; bromine, yellow; chlorine, green; oxygen, brown-red;
rhodium, red.
Orange needle-like crystals of 4 were grown upon slow evaporation of the mother
liquor (CH OH) at ambient temperature over a period of two days. Compound 4 is
3
crystallizing in the trigonal crystallographic system, P3 21 space group and the
2
molecular structure of 4 is presented in Fig. 5.
The coordination geometry is that of a distorted octahedron and the ligands around
the rhodium(III) center are in the mer configuration. In complex 4, the Rh–O(2) bond
length of the coordinated methanol is at 2.110(4) Å and compares well with that
reported for complex 3. In addition, the RhN(1) and RhN(2) bond distances are at
1
.974(4) Å and at 2.041 (4) Å respectively.
Moreover, it should be mentioned that the phenyl ring involving the atoms C15 to
C20 is disordered over two distinct positions in the ratio 1:1. This side occupancy is
omitted for clarity. Selected bond lengths and angles for complex 4 are listed in Table
2
.
2
2

Fig. 5. Molecular structure of 4. Hydrogen atoms are omitted for clarity. The ellipsoids
were plotted at 50% probability.
In the crystal structure of 4 pair of molecules display intermolecular  stacking
interactions between the quinoline and phenyl rings of symmetry related molecules
(distance between centroids of 3.604 Å and 3.611 Å). The dimeric subunits of 4 form
long chains that are separated by a  stacking at a distance of 3.475 Å (Fig. 6). These
are connected each other with typical O(1)H(1)∙∙∙Cl(3) hydrogen bonding interactions
at a distance of 3.197 (4) Å and a bond angle of 148.7°.
Fig. 6. Intermolecular  stacking interactions in complex 4. Carbon, gray; hydrogen,
light gray; nitrogen, blue; bromine, yellow; chlorine, green; oxygen, brown-red;
rhodium, red.
3
.5. Electronic spectra
The solution absorption spectra of the organic ligands 1, 2 and the complexes 3, 4
in DMSO are depicted in the supporting information (Fig. S18a). The intense high
2
3

energy bands at 279 nm for 1 and at 289 nm for 2 are assigned to π-π* transitions. For
, the bands centered at 343 and 358 nm are red shifted as compared to the Br
2
derivative 1 (bands at 327, 343 nm), apparently due to the hydroxyphenyl group that
acts as a better -donor.
The electronic spectra of 3 and 4 display typical metal to ligand charge transfer
bands (MLCT) at 342 nm and 355 nm for 3, while for complex 4 a rather broad MLCT
band centered at 392 nm with shoulders at 367 nm and 426 nm respectively, is observed.

4
6
Furthermore the spectra were recorded within the concentration range of 10 to 10

5
7
Μ (3) and 10 to 10 Μ (4), without changes in the spectral shape (Figs. S18b,c) and
the Lambert-Beer law is obeyed (linear correlation between absorption and
concentration). These results imply the absence of inter- or intra-molecular -
interactions in solution [47].
Complexes 3 and 4 are stable in DMSO solution, over a period of 24 h. No new
peaks or isosbestic points were observed (Figs. S19,S20). For complex 4, the absorption
spectrum shows a slow decrease of the band intensities without shifting of the
absorption maxima. This could be explained on the basis of progressive aggregation
that leads to a decrease of the complex concentration [48].
1
3
.6. Solution stability of 3 and 4 by H NMR spectroscopy
For the biological evaluation of this class of compounds their stability in the
required medium is a prerequisite. DMSO is a very common solvent used for this
purposes, and its coordination ability to the relevant metal center via either S or O atoms
has been well documented in the literature [46,495051]. Complexes 3, 4 were
dissolved in DMSO-d solution at room temperature and their stability was monitored
6
1
by H NMR spectroscopy, without exposure to direct light, over a period of 24 to 48 h.
In fact, no new signals were observed in the aromatic region (Fig. S21 (3), Fig. S22
(4)). Therefore, complexes 3 and 4 can be potentially used for the proposed biological
1
evaluation in this medium. Monitoring of the H NMR spectra of 3, 4 for a longer period
of time in the absence of light, reveals that slow isomerization to equilibrium mixtures
of the mer (3, 4) and the fac (3a, 4a) isomers takes place. In fact 3/3a (36/64) and 4/4a
(
50:50) isomers were formed after 9 and 5 days respectively in DMSO-d solution at
6
2
4

room temperature. This is in accord to the remarks of Sheldrick et al. on the structurally
related complexes mer-[RhCl L(DMSO)] (L = dpq, dppn), where a 92:8 mer/fac ratio
3
was found for L = dpq after 4 days [52].
3
.7. Biological evaluation
Herein, we report for the first time on the biological study of the organic precursors
1
and 2 and the corresponding octahedral rhodium(III) complexes 3 and 4, emphasizing
on the evaluation of the anti-platelet activity expressed.
IC (μM) towards
5
0
PAF (10 M, in the cuvette) in
10
Precursors and Rh Complexes
SD reference
WRPs
Ligands
1
2
3.4
4.4
6.4
32
0.2 this work
0.2 this work
L
L1
0.5
15
[20]
[14,18]
Starting materials
RhCl ×3H O
67
22
[14]
3
2
Complexes
2
5

1
mer-[Rh(L )Cl (CH OH)]
2.6
1.0
3.9
2.0
[14]
3
3
(3)
4)
0.6 this work
0.2 this work
(
The new substances were biologically evaluated towards PAF induced activities in
washed rabbit platelets (WRPs) aggregation assays. PAF is a link of an increasing list
of mediators implicated in physiological as well as pathological inflammatory
processes, since it has the ability to stimulate inflammatory cells and to mediate both
cell-cell interactions and intracellular signaling. On the other hand, PAF-inhibitors have
also anti-inflammatory activity [53]. In fact, the anti-inflammatory action of the new
compounds may be exerted by their interaction with PAFR, blocking the PAFR
activation. The inhibitory effect of all the compounds tested, including the rhodium
precursor RhCl ×H O, was expressed by their IC value in μM as presented in Table
3
2
50
3
. Table 3
Inhibition of PAF induced aggregation by the rhodium(III) coordination compounds,
the ligands and the rhodium precursor toward WRPs
The IC values reflect the inhibition strength of each compound, since a low IC
50
5
0
value reveals stronger inhibition of the PAF-induced aggregation for a given
concentration. As far as the biological behavior of the organic ligands is concerned,
valuable information is obtained upon comparison of the structurally related
compounds 1 and 2 with the organic precursor 2-(2'-pyridyl)quinoline (L) [20], that
served as a control compound (Scheme 6).
2
6

Scheme 6. Comparison of the organic precursors presented in this work, displaying (a)
pyridyl-quinoline or (b) pyridyl-quinoxaline core.
From this scheme it is evident that 1, 2 and L, all display the typical pyridyl-quinoline
1
2
core, differing however on the nature of the R , R substituents. Both 1 and 2 have the
1
1
same functional group R at the quinolinic part of the common core ( R = Ph), as
2
2
2
opposed to the R substituent that changes from 1 (R = –Br) to 2 (R = –PhOH)
respectively. On the other hand, the less sterically demanding ligand L contains the
1
2
1
2
same R , R groups (R = R = H).
The results listed in Table 3 clearly show that within the organic precursors tested,
1
is the most potent inhibitor. In fact, the IC value drops from 6.4 μM (L) to 4.4 μM
50
(2) and to 3.4 μM (1) respectively. It seems therefore that for this class of compounds,
the inhibitory effect against the PAF induced aggregation (anti-PAF activity) is affected
1
2
but to different extends by the presence of the R and R substituents. Presumably the
2
presence of the bulky R = –PhOH group, may be responsible for the weaker inhibitory
effect of 2 as compared to 1.
On the other hand, a plausible explanation about the less pronounced inhibitory
1
effect of L = 2-(2'-pyridyl)quinoxaline [15], in comparison to that of L, could be made
2
7

according to the aforementioned manner (Scheme 6b). From the obtained experimental
1
data, it is obvious that the inhibitory effect of L , comprising the quinoxaline moiety,
is reduced almost five times on going to L that holds the quinolinic functional group.
1
In fact, the IC dropsfrom 32 μM (L ) to 6.4 μM (L) respectively. The data (IC values
5
0
50
in Table 3) also suggest that, in the case of PAF-induced WRPs aggregation, the
octahedral rhodium(III) complexes 3, 4 exhibit a biological activity in the micromolar
range with IC values of 1 μM and 3.9 μM respectively. This is in agreement with the
5
0
inhibitory effect reported (IC50 = 2.6 μM) for the structurally related mer-
1
[
Rh(L )Cl (CH OH)] complex [15].
3
3
Upon a careful examination of the results listed above we further noticed, that
coordination of 1 to the rhodium(III) metal center, enhances the inhibitory potency
against the PAF-induced aggregation (synergetic effect). In fact, on going from 1 (3.4
μM) to 3 (1.0 μM) the PAF inhibitory effect increases approximately by a factor of
three, (the IC value decreases) rendering compound 3 more potent against PAF
5
0
inhibition. Complex 4 incorporating the bulkier organic precursor 2 shows a slightly
higher inhibitory activity (IC = 3.9 μM) than that of 2 (IC = 4.4 μM). From these
5
0
50
findings one may propose that the biological activity of 4 and of the organic precursor
practically remains unaltered. In any case, the combination of the rhodium precursor
RhCl ×3H O (IC = 67 μM) and 1 or 2 affording 3 and 4 (synergetic effect), leads to a
2
3
2
50
dramatic increase of the inhibitory activity against the PAF-induced aggregation.
Conclusively the inhibitory activity against PAF decreases in the following order: 3 >
1
> 4 > 2.
From the library of more than 30 metal-based inhibitors created so far [3,17], we
have realized that this effect (synergetic effect), is a general trend which is followed in
the present case as well. These results underline the substantial role of the final structure
of the proposed metal-based inhibitor. On the other hand, it should be noted that
parameters such as the total charge on the metal complex cannot be excluded [54]. The
neutral character of the rhodium(III) complexes 3, 4 is a feature that may affect
permeability through cell membranes and needs further investigation [55].
Notably the IC values of the title compounds 3, 4 are comparable or even better
5
0
than the inhibitory effect of Gingolides, a class of natural specific inhibitors from the
leaves and roots of the Chinese tree Gingo biloba. This becomes evident within the
2
8

series of Gingolides BN 52020 (IC = 3.6 μM), BN 52021 (IC = 9.7 μM) and BN
5
0
50
5
2022 (IC = 38 μM) respectively [56]. This is of particular interest denoting the great
50
potency of the present compounds, as possible anti-inflammatory agents in the future,
provided that they display the required pharmacological profile (toxicity evaluation
etc.). According to Leung, Ma and co-workers [10] this field is still immature and very
promising, aiming to produce a series of metal complexes that could be approved in the
treatment of inflammatory diseases.
Although the biological probe studied in this report is a complicated system, we
managed to confirm the preliminary structure-activity relationships established [3,17]
and as a result to increase the number of metal-based inhibitors of PAF (compounds
with anti-platelet potency) in the library of metal complexes that has been created.
Finally, the results of this study will help us to rational design new compounds with
improved anti-PAF activity. To reinforce our experimental findings further studies are
required, and towards this goal work is currently underway.
4
. Conclusions
In conclusion, we have successfully synthesized and characterized two new
substituted pyridyl-quinoline ligands (1, 2) that served as potent building blocks for the
synthesis of the octahedral rhodium(III) complexes (3, 4). The biological evaluation of
the new compounds 14 towards the PAF-induced aggregation in washed rabbit
platelets (WRPs) is also reported, revealing that all substances display potent activities.
Complex 3 bearing the less sterically demanding ligand 1 is a better inhibitor of PAF
as compared to complex 4. In total 3 and 4 are potent inhibitors of PAF in the
micromolar range (IC = 1.0 μM to 3.4 μM).
5
0
Notably it can be pointed out that the established structure-activity relationships
from our previously reported studies are followed as well, for this type of metal
complexes (same trend). Thus the nature of both the metal center and of the organic
ligand, contribute substantially to the biological action expressed (synergetic effect).
Moreover the size of the metal complex, determines the extent of their inhibitory action.
Conclusively, the new substances could be successfully added to the existing
library of more than 30 small molecules with potent anti-PAF activity. Accordingly,
these findings would enable us to estimate or even to predict the anti-PAF activity of
2
9