Structure-specific adipogenic capacity of novel, well-defined ternary Zn(II)-Schiff base materials. Biomolecular correlations in zinc-induced differentiation of 3T3-L1 pre-adipocytes to adipocytes

Article abstract of DOI:10.1016/j.jinorgbio.2015.08.014

Among the various roles of zinc discovered to date, its exogenous activity as an insulin mimetic agent stands as a contemporary challenge currently under investigation and a goal to pursue in the form of a metallodrug against type 2 Diabetes Mellitus. Poi

Full text of DOI:10.1016/j.jinorgbio.2015.08.014

Structure-Specific Adipogenic Capacity of Novel, Well-defined Ternary
Zn(II)-Schiff Base Materials. Biomolecular Correlations in Zinc-Induced
Differentiation of 3 T3-L1 Pre-adipocytes to Adipocytes
O. Tsave, E. Halevas, M.P. Yavropoulou, A. Kosmidis Papadimitriou,
J.G. Yovos, A. Hatzidimitriou, C. Gabriel, V. Psycharis, A. Salifoglou
PII:
DOI:
Reference:
S0162-0134(15)30057-X
doi: 10.1016/j.jinorgbio.2015.08.014
JIB 9785
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
16 June 2015
2 August 2015
5 August 2015
Please cite this article as: O. Tsave, E. Halevas, M.P. Yavropoulou, A. Kosmidis
Papadimitriou, J.G. Yovos, A. Hatzidimitriou, C. Gabriel, V. Psycharis, A. Sali-
foglou, Structure-Specific Adipogenic Capacity of Novel, Well-defined Ternary Zn(II)-
Schiff Base Materials. Biomolecular Correlations in Zinc-Induced Differentiation of
3 T3-L1 Pre-adipocytes to Adipocytes, Journal of Inorganic Biochemistry (2015), doi:
10.1016/j.jinorgbio.2015.08.014
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.

ACCEPTED MANUSCRIPT
1
Structure-Specific Adipogenic Capacity of Novel, Well-defined Ternary Zn(II)-Schiff
Base Materials. Biomolecular Correlations in Zinc-Induced Differentiation of 3T3-L1
Pre-adipocytes to Adipocytes.
O. Tsave,a E. Halevas,a M. P. Yavropoulou,d A. Kosmidis Papadimitriou,a J. G. Yovos,d A.
Hatzidimitriou,b C. Gabriel,a,e V. Psycharis,c A. Salifogloua*
a
Laboratory of Inorganic Chemistry, Department of Chemical Engineering, Aristotle
University of Thessaloniki, Thessaloniki 54124, Greece
b
Laboratory of Inorganic Chemistry, Department of Chemistry, Aristotle University of
Thessaloniki, Thessaloniki 54124, Greece
c
Institute of Advanced Materials, Physicochemical Processes, Nanotechnology and
Microsystems, Department of Materials Science, N.C.S.R. ―Demokritos‖, Aghia Paraskevi
15310, Attiki, Greece
d
Laboratory of Clinical and Molecular Endocrinology, 1st Department of Internal Medicine,
AHEPA, University Hospital, Aristotle University of Thessaloniki, Thessaloniki 54124,
Greece
e
Center for Research of the Structure of Matter, Magnetic Resonance Laboratory,
Department of Chemical Engineering, Aristotle University of Thessaloniki, Thessaloniki
54124, Greece
*
Author to whom correspondence should be addressed.
Tel: +30-2310-996-179 Fax: +30-2310-996-196 E-mail: salif@auth.gr

ACCEPTED MANUSCRIPT
2
Abstract
Among the various roles of zinc discovered to date, its exogenous activity as an insulin
mimetic agent stands as a contemporary challenge currently under investigation and a goal to
pursue in the form of a metallodrug against type 2 Diabetes mellitus. Poised to investigate the
adipogenic potential of Zn(II) and appropriately configure its coordination sphere into well-
defined anti-diabetic forms, a) a series of new well-defined ternary dinuclear Zn(II)-L (L =
Schiff base ligands with a variable number of alcoholic moieties) compounds were
synthesized and physicochemically characterized, c) their cytotoxicity and migration effect(s)
in both pre- and mature adipocytes was assessed, d) their ability to effectively induce cell
differentiation of 3T3-L1 pre-adipocytes into mature adipocytes was established, and d)
closely linked molecular targets involving or influenced by the specific Zn(II) forms were
perused through molecular biological techniques, cumulatively delineating factors involved in
Zn(II)-induced adipogenesis. Collectively, the results a) reveal the significance of key
structural features of Schiff ligands coordinated to Zn(II), thereby influencing its (a)toxicity
behavior and insulin-like activity, b) project molecular targets influenced by the specific
forms of Zn(II) formulating its adipogenic potential, and c) exemplify the interwoven
relationship between Zn(II)-L structural speciation and insulin mimetic biological activity,
thereby suggesting ways of fine tuning structure-specific zinc-induced adipogenicity in future
efficient antidiabetic drugs.
Keywords: Schiff ligand-based zinc reactivity, zinc metallodrugs, structure-specific
bioactivity selection, fine-tuning zinc insulin-like activity, zinc-induced adipogenesis,
biomolecular correlations

ACCEPTED MANUSCRIPT
3
1. Introduction
Diabetes Mellitus (DM) is a group of metabolic diseases characterized by hyperglycemia due
to absence of insulin or impaired insulin secretion. Chronic hyperglycemia in diabetes is
associated with long-term micro and macrovascular complications, significantly increasing
mortality and morbidity of diabetic patients. DM is generally classified into two types: a)
type 1 or insulin-dependent type, caused by autoimmune destruction of pancreatic beta cells,
and b) type 2 or non-insulin dependent, caused by the dysfunction of beta cells and insulin
resistance, which is due to the inability of insulin target tissues to respond properly to insulin
[1-5]. Insulin action involves an intracellular signaling cascade, initiated by insulin binding to
its trans-membrane receptor, resulting in receptor autophosphorylation, through activation of
the receptor tyrosine kinase, and phosphorylation of insulin receptor substrates (IRSs). In
turn, phosphorylated IRS activates downstream kinases, such as phosphatidylinositol 3-kinase
(PI3K) and Akt kinase, both essential for insulin-induced glucose transport in the periphery.
Although the underlying mechanisms of insulin resistance are not completely understood,
impaired insulin-dependent PI3K activation and downstream signaling is considered the
convergence point of insulin dysfunction. Insulin resistance of peripheral target tissues is
caused by a reduced number of functional glucose transporters (GLUTs), mainly GLUT 4. In
murine cells, the defect appears to emerge through failure of bound vesicle carriers to displace
the cytoplasmic membrane in response to insulin. In adipocytes the displacement is also
limited, but the main mechanism of insulin resistance in these cells is the pre-translational
depletion of mRNA for GLUT 4 transporters [6,7].
Zinc is an important trace element critical to the function of over 300 enzymes [8,9] and plays
an important role in processes like DNA/RNA synthesis, cell division and apoptosis, insulin
synthesis, secretion and signaling [10-14]. In addition, zinc has been linked to DM as part of
the insulin complex [15], and several studies have investigated the role of zinc metabolism in
diabetic animals and humans. Zinc appears to play a role in modulating insulin receptor
tyrosine kinase activity in the skeletal muscle of a genetic DM type 2 model mouse, similar to
the action of vanadium [16]. Moreover, zinc was proposed to affect carbohydrate metabolism
through the insulin receptor, PTP1B, and other related proteins [17]. In fact, zinc and DM
converge at several points during metabolism in cells [18,19]. In 1980, zinc was found to
stimulate rat adipocyte lipogenesis in vitro, similar to the action of insulin [20], and exert
antidiabetic effects in vivo as orally given ZnCl2 in rats with streptozotocin-induced diabetes
and leptin-deficient mice (ob/ob mice) [21,22]. As the bioavailability of ZnCl2 is relatively
low, the coordination chemistry of Zn(II) ion was explored, and the first orally active insulin-

ACCEPTED MANUSCRIPT
4
mimetic and antidiabetic Zn(II)–picolinato complexes were proposed in 2002 [23]. Since
then, several Zn(II) complexes with different coordination structures have been synthesized
[18-24].
In an effort to identify new anti-diabetic drugs containing Zn(II) and fulfilling the basic
criteria of a) atoxic behavior, and b) optimal insulin mimetic activity, due attention was paid
to well-defined organic substrates, which appear to play a major role in formulating Zn(II)
insulin mimetic activity. They, thus, attracted our attention as potential Zn(II) binders [25].
Among such organic reagents, Schiff bases a) possess structural similarities with natural
biological substances, b) are synthesized through relatively simple preparation procedures,
and c) present valuable structural flexibility, enabling design of suitable (bio)chemical
reactivity properties [26,27]. They have been widely studied [28] because of their role as
intermediates in a number of enzymatic reactions and non-enzymatic glycosylations [29], and
their industrial, antifungal, antibacterial, anticancer, antidiabetic, antiviral and herbicidal
applications [30,31,32,33,34,35].
Guided by the idea that properly designed Schiff bases may promote Zn(II) binding, thereby
giving rise to bioactive ternary Zn(II)-L-(L’) insulin mimetic compounds [30,31,32] capable
of reducing blood glucose levels, efforts were launched to a) design and synthesize well-
defined ternary Zn(II)-Schiff base compounds of specific structure (organic substrates bearing
systematically varying number of alcohol moieties attached to a vanillin core) and
characterize them physicochemically, b) investigate their ability to induce 3T3-L1 fibroblast
to mature adipocyte differentiation, c) examine the insulin mimetic potency of such
compounds and determine in vitro their arising biological activity profile(s), d) investigate the
interactions of such molecules with specific molecular targets in the insulin-induced glucose
catabolic pathways known to date, and e) correlate chemical and structural features of the
ternary Zn(II) compounds with the observed activities at the cellular level, thereby
formulating biological activity profile-physicochemical factor relationships that enable
development of optimal Zn(II) insulin mimetic drugs. To this end, we report herein the
synthesis and physicochemical characterization of a series of newly designed ternary Zn(II)-
Schiff base compounds, which are systematically studied with respect to their toxicity and
potential to induce cell differentiation.
2. Experimental Section
2.1. Materials and Methods. All experiments were carried out under aerobic conditions. The
following starting materials were purchased from commercial sources (Sigma, Fluka) and
were used without further purification:
tromethamine, 2-amino-1,3-propanediol,

ACCEPTED MANUSCRIPT
5
ethanolamine, o-vanillin, and zinc acetate dihydrate (Zn(CH3COO)2 · 2H2O). Solvents:
ultrapure water, methanol, ethanol, diethyl ether. The compounds isolated and dried under
vacuum at room temperature are air-stable.
2.2. Physical measurements. FT-Infrared spectra were recorded on a Perkin Elmer 1760X
FT-infrared spectrometer. A ThermoFinnigan Flash EA 1112 CHNS elemental analyser was
used for the simultaneous determination of carbon, hydrogen, and nitrogen (%). The analyser
operation is based on the dynamic flash combustion of the sample (at 1800°C) followed by
reduction, trapping, complete GC separation and detection of the products. The instrument is
a) fully automated and controlled by PC via the Eager 300 dedicated software, and b) capable
of handling solid, liquid or gaseous substances.
13
2.2.1. Solid-State NMR spectroscopy. High resolution solid-state CP-MAS C NMR spectra
for 1-3 were obtained on a Varian 600 MHz spectrometer operating at 150.86 MHz. In each
case, a sufficient sample quantity was placed in a double resonance HX probe with a 3.2 mm
rotor. The spinning rate used was 15.0 kHz at 25 °C. The RAMP-CP pulse sequence of the
VnmrJ library was applied, whereby the 13C spin-lock amplitude is varied linearly during CP,
while the 1H spin-lock amplitude is kept constant. RAMP-CP eliminates the Hartmann-Hahn
matching profile dependence from the MAS spinning rate and optimizes signal intensity. The
solid-state spectra were a result of the accumulation of 512 scans. The recycle delay used was
2s, the 90° pulse was 2.0 μs, and the contact time was 3.0 ms. The CH signal of adamantane
was used as an external reference (37.78 ppm with respect to TMS).
1
2.2.2. Solution NMR Spectroscopy. Solution H- and 13C-NMR experiments for 1 and 3,
were carried out on Varian 600 MHz spectrometer. The sample concentration was ~5 mM.
Freshly prepared compounds were dissolved in D2O. Due to solubility problems for 2 at
1
13
concentrations up to ~5 mM, well-resolved H- and C-NMR spectra could not be obtained.
Carbon spectra were acquired with 5000 transients, a spectral width of 37000 Hz and a
relaxation delay of 2s. Proton spectra were acquired with 128 transients and a spectral width
of 9000 Hz. Experimental data were processed using VNMR routines. Spectra were zero-
filled and subjected to exponential apodization prior to FT. Chemical shifts () are reported
in ppm, while spectra were referenced by the standard experimental setup.
2.2.3. Thermal Studies. A TA Instruments, model Q 600, system was used to run the
simultaneous TGA experiments. The instrument mass precision is 0.1 μg. About 10 mg of
each sample was placed in an open alumina sample pan for each experiment. A mixture of
helium and high purity air (80/20 in N2/O2) was used at a constant flow rate of 100 mL/min,
depending on the conditions required for running the experiments. During the experiments,

ACCEPTED MANUSCRIPT
6
the sample weight loss and rate of weight loss were recorded continuously under dynamic
ο
conditions, as a function of time or temperature, in the range 45–550 C. Prior to activating
the heating routine program, the entire system was purged with the appropriate gas for 10
min, at a rate of 400 mL/min, to ensure that the desired environment was established.
2.3. Preparation of compounds
2.3.1. Preparation of ligands LiH (i=1-3). Ligands L1H, L2H and L3H were prepared
according to the literature [36,37,38].
2.3.2. Preparation of complex Zn2(L1)(CH3COO)3 (1). To a solution of o-vanillin (0.15 g, 1.0
mmol) in EtOH (10 mL) tromethamine (0.12 g, 1.0 mmol) was added under stirring. The
resulting yellow solution was refluxed for two hours under continuous stirring and then
.
cooled to room temperature. Subsequently, a solution of Zn(CH3COO)2 2H2O (0.44 g, 2.0
mmol) in EtOH (10 mL) was added under continuous stirring. The resulting clear, yellowish
reaction mixture was refluxed for an additional 2 h and then cooled to room temperature.
Subsequently, the reaction flask was placed at 4 oC and diethyl ether was added. Two weeks
later, yellow crystalline material precipitated at the bottom of the flask. The product was
isolated by filtration and dried in vacuo. Yield: 0.34 g (61%). Anal. Calcd for 1,
Zn2(L1)(CH3COO)3 (1). (C18H25NO11Zn2 Mr 562.13): C, 38.46; H, 4.48; N 2.49. Found: C,
38.38; H, 4.33; N 2.44.
2.3.3. Preparation of complex [Zn2(L2)(CH3COO)3].H2O (2). To a solution of o-vanillin
(0.15 g, 1.0 mmol) in EtOH (10 mL) 2-amino-1,3-propanediol (0.091 g, 1.0 mmol) was added
and the resulting yellow solution was refluxed for 2 h under continuous stirring. Then, it was
.
cooled down to room temperature. Subsequently, a solution of Zn(CH3COO)2 2H2O (0.22 g,
1.0 mmol) in EtOH (10 mL) was added under continuous stirring and the resulting clear,
yellowish solution was refluxed for an additional 2 h. Then, it was cooled down to room
o
temperature. Yellow crystals of the complex were grown by slow evaporation at 4 C. The
product was isolated by filtration and dried in vacuo. Yield: 0.32 g (58%). Anal. Calcd for 2,
[Zn2(L2)(CH3COO)3].H2O (2). (C17H25NO11Zn2, Mr 550.12): C, 37.12; H, 4.58; N 2.55.
Found: C, 36.98; H, 4.43; N 2.24.
2.3.4. Preparation of complex Zn2(L3)2(CH3COO)2 (3). To a solution of o-vanillin (0.15 g,
1.0 mmol) in EtOH (10 mL) ethanolamine (0.060 mL, 1.0 mmol) was added and the resulting
yellow solution was refluxed for 2 h. Then, it was cooled down to room temperature.
Subsequently, a solution of Zn(CH3COO)2 · 2H2O (0.22 g, 1.0 mmol) in EtOH (10 mL) was
added under continuous stirring and the resulting clear, yellowish solution was refluxed for an
additional 2 h. Then, it was cooled down to room temperature. Subsequently, the reaction

ACCEPTED MANUSCRIPT
7
o
flask was placed at 4 C and diethyl ether was added. Two weeks later, yellow crystalline
material precipitated at the bottom of the flask. The product was isolated by filtration and
dried in vacuo. Yield: 0.22 g (35%). Anal. Calcd for 3, Zn2(L3)2(CH3COO)2 (3).
(C24H30N2O10Zn2, Mr 637.27): C, 45.23; H, 4.75; N 4.40. Found: C, 45.03; H, 4.64; N 4.34.
2.4. X-ray crystal structure determination. X-ray quality crystals of compounds 1 and 3
were grown from a mixture of ethanol-diethyl ether, whereas crystals of compound 2 were
grown from ethanol solutions. A crystal of 1 (0.15  0.30  0.46 mm) and 2 (0.07  0.09 
0.15 mm) were taken from the mother liquor and immediately cooled to -113 oC. Diffraction
measurements were made on a Rigaku R-AXIS SPIDER Image Plate diffractometer using
graphite monochromated Cu Kα radiation. Data collection (ω-scans) and processing (cell
refinement, data reduction and empirical absorption correction) were performed using the
CrystalClear program package [39]. The structures were solved by direct methods using
SHELXS-97 and refined by full-matrix least-squares methods on F2 with SHELXL-97 [40].
In 1, one of the terminal alcoholic oxygen atoms was found disordered and was refined
anisotropically over two positions, O(5) and O(5a), with occupancy factors summing one. In
both 1 and 2, hydrogen atoms were either located by difference maps and refined isotropically
or introduced at calculated positions as riding on bonded atoms.
A crystal of 3 (0.06  0.31  0.33 mm) was taken from the mother liquor and mounted at
room temperature on a Bruker Kappa 2 APEX diffractometer equipped with a triumph
monochromator using Mo K radiation. Unit cell dimensions were determined and refined
by using the angular settings of at least 100 high intensity reflections (>10 (I)) in the range
o
19<2<40 . Intensity data were recorded using and scans. All crystals presented no
decay during data collection. The collected frames were integrated with the Bruker SAINT
software package [41] using a narrow-frame algorithm. Data were corrected for absorption
using the numerical method (SADABS) based on crystal dimensions [42]. Data refinement
(full-matrix least-squares methods on F2) and all subsequent calculations were performed
using the Crystals version 14.40b program package [43]. The structure was solved by
SUPERFLIP method [44]. Molecular illustrations with 50% ellipsoids probability were
drawn by the Diamond 3.1 crystallographic package [45]. In all three structures (1-3), all
non-hydrogen atoms were refined anisotropically. Crystallographic details for 1, 2 and 3 are
summarized in Table 1. Further experimental crystallographic details for 1: 2θmax=133o;
number of reflections collected/unique/used, 23907/3653 [R(int) = 0.0372]/3653; 387
parameters refined; Δ/σ = 0.000; (Δρ)max/(Δρ)min = 0.448/-0.615 e/Å3; R/Rw (for all data),

ACCEPTED MANUSCRIPT
8
0.0327/0.0784. Further experimental crystallographic details for 2: 2θmax=130o; number of
reflections collected/unique/used, 23117/3577[R(int) = 0.0420]/3577; 341 parameters refined;
Δ/σ = 0.002; (Δρ)max/(Δρ)min = 0.580/-0.419 e/Å3; R/Rw (for all data), 0.0297/0.0746. Further
experimental crystallographic details for 3: 2θmax=56.2o; number of reflections
collected/unique/used, 50189/6807/3666 [R(int) = 0.0438]/3666; (Δρ)max/(Δρ)min = 0.88/-0.63
e/Å3; R/Rw (for all data), 0.0547/0.0764.
2.5. Cell cultures and biological tests
2.5.1. Cell culture
Both pre- and mature adipocytes (3T3-L1) were cultured in 75 cm2 cell culture flasks under
o
appropriate conditions (5% CO2 at 37 C and standard humidity) in Dulbecco’s modified
Eagle’s medium DMEM (Gibco), supplemented with 10% Fetal Bovine Serum (FBS) and 1%
penicillin-streptomycin.
2.5.2. Induction of adipogenesis
3T3-L1 fibroblasts were differentiated into mature adipocytes according to a standard
differentiation protocol [46]. Two days (day 0) post 70% confluency, cells were treated with
0.5 mM 3-isobutyl-1-methylxanthine (Sigma), 1.0 μΜ dexamethazone (Sigma) and 10 ng/mL
of insulin as final concentrations in DMEM. Two days later (day 2), the cells were treated
only with insulin and the medium was changed every two days. All experiments were
performed on the 8th day of the differentiation process.
2.5.3. Cell viability assay
In an effort to assess cell survival rates post-stimulation with the newly synthesized zinc
compounds, both 3T3-L1 pre-adipocytes and mature adipocytes differentiated with insulin as
mentioned above, were seeded in 96-multi-well plates (2500 cells/well), and incubated with
the title zinc compounds for 24 and 48 h in DMEM. The assay relies on the quantitation of
ATP present, thereby attesting to the presence of metabolically active cells. The reagent is
added to the cell culture (ratio 1:1) without removing the supernatant. The luciferase-
produced luminescence signal intensity is determined with a Glomax 96 microplate
luminometer (Promega Corporation, WI, USA).
2.5.4. Induction of adipogenesis with Zn(II)-compounds
3T3-L1 pre-adipocytes were differentiated into mature adipocytes with respect to a standard
differentiation protocol (vide supra). Pre-adipocytes were treated with either 10 ng/mL of
insulin and\or selected well-defined zinc compounds (1-25 μΜ). The tested compounds were
dissolved in DMEM (10% FBS, 1% penicillin/streptomycin), followed by sterile filtration. In
all experiments run, insulin-treated cells were used as positive control. A control group with

ACCEPTED MANUSCRIPT
9
no treated cells (without insulin or zinc compounds) was also included. On the 8th day of the
differentiation process, cell differentiation was assessed with oil red O staining and Real-
Time PCR for relative expression of PPAR-γ, GLUT 1, GLUT 3 and GLUT 4. All tests were
carried out in triplicate.
2.5.5. RT-PCR assay
Total RNA was extracted from cells on the 8th day of the differentiation protocol, using Trizol
reagent (LifeScience, Chemilab). Synthesis of cDNA was performed with the iScript cDNA
synthesis kit (BioRad), according to manufacturer instructions. RT-PCR was run on Rotor
Gene Q (Qiagen) using the QuantiTect SYBR Green PCR kit and appropriate reagents
(Qiagen).
2.5.6. Cell migration assay
The inhibition of the endogenous migrating capacity of 3T3-L1 pre-adipocytes was assessed
through an in vitro scratch assay. To this end, cells were seeded in 35 mm cell culture dishes
in DMEM. When cells reached 70-80% confluency, a) a scratch in the monolayer traversing
the entire diameter of each culture dish was engraved using a sterile pipette tip, and b) cells
were allowed to grow in the culture medium in the presence of a final concentration of 25 μΜ
of the title Zn(II) compounds. Cells were visualized using an Axio Observer Z1 microscope,
with a 10x phase contrast water immersion objective (Carl Zeiss, GmbH Lena, Germany).
Images were captured, using an AxioCam Hc camera, at two distinct time points 24 and 48 h
after the scratch had been made.
2.5.7. Statistical analysis
The data were presented as average and standard error mean (SEM) values of triplicate sets of
independent measurements. Mean survival rates and SEMs were calculated for each group.
Absolute survival rates were calculated for each control group and one way analysis of
variance (ANOVA) was performed for all pair comparisons, followed by post hoc analyses
(Tukey) and Dunnet in the case of RT-PCR experiments. Degrees of significance were
assessed by three different rating values: *p<0.05 (significant), **p<0.01 (highly significant)
and ***p<0.001 (extremely significant) or non-significant (p>0.05).
3. Results
3.1. Synthesis. The synthetic exploration of the binary Zn(II)-Schiff base systems was based
on the following Schiff base ligands L1H-L3H, synthesized from o-vanillin and tromethamine
(L1H), 2-amino-1,3-propanediol (L2H), and ethanolamine (L3H) [36-38], respectively.
Subsequently, the Zn2(L1)(CH3COO)3 (1) complex was synthesized in a facile fashion from
simple reagents in alcoholic solutions. In a typical reaction, Zn(CH3COO)2·2H2O reacted

ACCEPTED MANUSCRIPT
10
with Schiff base L1H using ethanol as a solvent. The overall stoichiometric reaction leading
to 1 is shown schematically below:
OMe
OH
EtOH
.
2 Zn(CH3COO)2 H2O
+
OH
OH
N
L1H
OH
[Zn2(L1)(CH3COO)3] + CH3COOH + 2 H2O
In a similar reaction, Zn(CH3COO)2·2H2O reacted with Schiff base L2H in ethanol. The
stoichiometric reaction leading to the formation of compound [Zn2(L2)(CH3COO)3].H2O (2) is
shown below:
OMe
OH
EtOH
.
2 Zn(CH3COO)2 H2O
+
OH
H
N
L2H
OH
[Zn2(L2)(CH3COO)3]
+
CH3COOH + 2 H2O
In an analogous reaction, Zn(CH3COO)2·2H2O reacted with Schiff base L3H in ethanol. The
stoichiometric reaction leading to the formation of Zn2(L3)2(CH3COO)2 (3) is shown below:
OMe
OH
EtOH
.
2 Zn(CH3COO)2 H2O + 2
OH
N
L3H
[Zn2(L3)2(CH3COO)2]
+
2 CH3COOH + 2 H2O
Diethyl ether, was added as a precipitating solvent to the reaction mixture in the first and third
reactions described above. The second reaction mixture was left to evaporate slowly. Yellow
crystalline materials emerged in all three reactions described above, the analytical
composition of which was consistent with the formulation in 1, 2 and 3, respectively (vide
supra). Positive identification of the crystalline products was achieved by elemental analysis,

ACCEPTED MANUSCRIPT
11
FT-IR spectroscopic methods and X-ray crystallographic determination for isolated single
crystals from 1-3. All complexes are stable in the crystalline form, in the air, for fairly long
periods of time. All three freshly prepared species are readily dissolved in water and DMSO,
sparingly soluble in alcohols (CH3OH, i-PrOH), and insoluble in acetone, acetonitrile and
dichloromethane at room temperature.
3.2. Description of X-ray Crystallographic Structures
The X-ray crystal structures of 1, 2, and 3 reveal discrete solid state lattices. The molecular
structures of 1-3 are given in Figs. 1A, 2A and 3, respectively; selected bond distances and
angles are listed in Table 2. The molecular structure of 1 and 2 consists of two Zn(II) ions,
–
–
one L– (L1 , L2 , respectively) and three acetato ligands, whereas the molecular structure of 3
–
consists of two Zn(II) ions, two L– (L3 ) and two acetato ligands. Therefore, all three
complexes 1-3 are dinuclear Zn(II) species. 1 crystallizes in the monoclinic space group
P21/n, whereas 2 crystallizes in the orthorhombic space group P21cn, both with four
molecules in the unit cell. In the first two complexes, the metal ions are bridged through the
deprotonated phenolato oxygen atom of the LH ligands, which acts as a monoatomic bridge,
and through two acetato ligands, which present the usual syn,syn bridging mode. The third
acetato ligand is coordinated to one of the metal ions in an asymmetric chelating mode. All
–
–
–
types of L ligands, i.e. L1 , L2 , L3 , in 1-3 respectively, act as tetradentate chelators bound
through the Ophen, Oalk, Nimi and Omet anchors. The two Zn(II) ions in 1-3 present different
coordination environments. One metal ion is coordinated to the Ophen and Omet anchors of the
L ligands and to four carboxylato oxygen atoms from the three acetate ligands, leading to a
distorted octahedral geometry. The second Zn(II) ion in 1-3 is five-coordinate, bound to the
Ophen, Nimi and Oalk anchors of the L– ligands and to two carboxylato oxygen atoms from the
two syn,syn acetato ligands. The coordination geometry of the five-coordinate metal ions is
best described as a trigonal bipyramidal, based on the value of the trigonality index, τ (0.70
for 1, 0.49 for 2, and 0.62 for 3). The Zn-Ophe, Zn-Nimi, Zn-Oacetate and C-Oacetate distances
were found to be consistent with those reported in the literature [47,48,49,50]. A Zn-Omet
distance of 2.385, 2.402 Å and 2.329 Å for Zn-OCH3 binding in 1, 2 and 3 respectively, is
attributable to weak binding. It has been observed on (O,N)-bound Zn(II) complexes reported
in the literature that distances in the range from 2.34 to 2.91 Å lead to highly distorted
geometries [51,52,53,54,55]. These Zn(II)-O distances are similar to the corresponding
distances in the Zn(II)-ATP complex [51], containing a Zn(II)-Ophosphato distance of 2.71 Å
(distorted octahedral), in Zn(II)-histidinato complexes [52,53], with Z(II)-Ocarboxylato distances
of 2.91 and 2.79 Å (distorted octahedral), in the Zn(II)-triazacyclododecane complex [54],

ACCEPTED MANUSCRIPT
12
with Zn(II)-Operchlorato distances of 2.528 and 2.818 Å (distorted trigonal bipyramidal), and in
the case of Zn(II)-crown ether complex [55], with Zn(II)-Ocrownether distances of 2.34, 2.56 and
2.60 Å. Of the three –CH2OH groups in 1 and the two –CH2OH groups in 2, only one is
bound to the five coordinate Zn(II) center, whereas the other –CH2OH groups remain free.
All these -OH groups are involved in hydrogen-bonding interactions in the lattice. Each
molecule in the lattice of 1 interacts with four neighboring molecules via four intermolecular
hydrogen bonds between the free -CH2OH groups defined by O(3) and O(4), which act as
donors and carboxylato atoms O(16) and O(13), respectively, which act as acceptors (Table
3). In doing so, the dinuclear complexes are linked to form layers extending parallel to the ac
plane (Fig. 1B). Two neighboring layers are also linked through intermolecular hydrogen
bonds between the third free -CH2OH group defined by the disordered O(5) atom, which acts
as a donor and the carboxylato atom O(16), which acts as an acceptor, thus forming double
layers (Fig. 1C). In the crystal lattice of 2, each Zn(II) complex interacts with two
neighboring molecules through two intermolecular hydrogen bonds between the free -CH2OH
group defined by O(14), which acts as a donor and the carboxylato oxygen O(5), which acts
as acceptor, thereby forming 1D chains parallel to the [110] crystallographic direction (Fig.
2B, Table 3). These chains form layers parallel to the (001) crystallographic plane stacked
along the c-axis and further linked through intermolecular hydrogen bonds involving the
lattice water molecules and the free -CH2OH groups of the ligand. Consequently, an overall
3D network (Fig. 2C) is generated. Successive layers of chains along the c-axis are shown in
magenta and blue color in Fig. 2C.
Compound 3 crystallizes in the monoclinic space group P21/n with four independent
molecules in the unit cell. In 3, two ligands act as tetradentate and tridentate binders,
respectively, the first through the Ophen, Oalk, Nimi and Omet anchors and the second through the
Ophen, Oalk and Nimi anchors, collectively giving rise to a distorted octahedral Zn(II) center and
a Zn(II) center with a distorted trigonal bipyramidal geometry (Fig. 3). The six coordinate
Zn(II) center is bound to the Ophen, Oalk and Nimi groups of the first ligand, the Ophen, Omet
terminals of the second ligand, with one coordination site being occupied by an oxygen atom
of one acetato group. The five coordinate Zn(II) center is bound to the Ophen, Oalk and Nimi
moieties of the second ligand, with two coordination sites being occupied by one carboxylato
oxygen from each of the two acetato groups. One of the acetate groups acts as a bidentate
bridge of the two Zn(II) centers. Two intramolecular hydrogen bond interactions occur in the
lattice of 3. The Zn(II)-bound –CH2OH group of the six coordinate Zn(II) center is involved
in a hydrogen bond interaction with the oxygen of the acetato group of the five coordinate

ACCEPTED MANUSCRIPT
13
Zn(II) center. The Zn(II)-bound –CH2OH group of the five coordinate Zn(II) center forms a
hydrogen bond with the second oxygen of the same acetato group.
3.3. FT-IR spectroscopy. The spectra of compounds 1-3 show active ν(OH) and ν(C=N)
vibrations at 3380-3439 and 1617-1644 cm-1, respectively, with the binding of the imine
nitrogen confirmed by lowering of the resonance frequency (13-15 cm-1). Specifically, ν(OH)
and ν(C=N) emerged at 3439 cm-1and 1628 cm-1 for 1, 3381 cm-1 and 1618 cm-1 for 2, 3431
cm-1 and 1644 cm-1 for 3. The νasym and νsym vibrations of the acetato groups were observed
in all cases. Specifically, the antisymmetric stretching vibrations νas(COO–) emerged at 1595
cm-1 for 1, 1562 cm-1 for 2, and 1582 cm-1 for 3. Symmetric stretching vibrations νs(COO–)
for the same group appeared at 1442 cm-1 for 1, 1448 cm-1for 2, and at 1443 cm-1 for 3, with
Δν being 114-153 cm-1. The frequencies of the observed carbonyl vibrations were shifted to
lower values in comparison to the corresponding vibrations in free acetate, indicating changes
in the vibrational status of the ligands upon binding to the Zn(II) ion. All data were further
confirmed by the X-ray crystal structures of 1-3.
3.4. Solid-state NMR spectroscopy. The 13C CP-MAS-NMR spectrum of 1 (Fig. 4A) was
consistent with the coordination mode of the Schiff base ligand bound to Zn(II) ion.
Specifically, the spectrum shows separate resonances for the various carbons of the Schiff
base ligand and the acetate moiety and exhibits a distinctly different pattern compared to the
free Schiff base. The resonances in the high-field region (19.6-22.6 ppm) could be assigned
to the methyl groups of the variably bound acetates. The resonance at 54.3 ppm is reasonably
ascribed to the methoxy group carbon atom. The signals in the range 61.0-65.7 ppm could be
ascribed to the carbons of the tromethamine tether. Αnother group of resonances in the range
115.8-151.6 ppm could be assigned to the benzene ring of the Schiff base. In the low field
region, a signal around 174.1 ppm could be assigned to the azomethine group carbon, with the
broad signal around 180.7 ppm being attributable to the carboxylato group carbons of the
acetates bound to Zn(II) [56]. The 13C CP-MAS-NMR spectrum of 2 (Fig. 4B) shows
resonances in the high-field region (20.5-23.1 ppm) assignable to the methyl groups of the
variably bound acetates. The resonance at 53.7 ppm is reasonably ascribed to the methoxy
group carbon atom. The signals in the range 59.9-77.7 ppm could be ascribed to the carbons
of the aminoalcohol tether. The group of resonances in the region 116.5-152.6 ppm could be
assigned to the presence of the benzene ring of the Schiff base ligand. In the low field region,
a signal around 170.9 ppm could be assigned to the azomethine group carbon atom, with the
broad signal around 180.1 ppm being attributable to the carboxylato group carbons of acetates
bound to Zn(II). The 13C CP-MAS-NMR spectrum of 3 (Fig. 4C) exhibits a distinctly

ACCEPTED MANUSCRIPT
14
different pattern of signals from 1 and 2. The resonance in the high-field region (22.1 ppm)
could be assigned to the methyl groups of the bound acetates. The resonance at 54.5 ppm is
reasonably ascribed to the methoxy group carbon atom. The signals in the range 56.7-78.3
ppm could be ascribed to the carbons of the aminoalcohol tether. The group of resonances at
111.5-162.3 ppm could be assigned to the presence of the benzene ring in the Schiff base. In
the low field region, a signal around 170.6 ppm could be assigned to the azomethine group
carbon atom, with the broad signal in the range 172.8-177.7 ppm being attributable to the
carboxylato group carbons of the acetates bound to Zn(II). All of the resonances appear to be
shifted to slightly higher fields compared to the free ligands. These observations are
consistent with the structure of the compounds revealed by X-ray crystallography.
3.5. Solution NMR spectroscopy. The solution 13C-NMR spectra of freshly prepared
compounds 1 and 3 were measured in D2O. The spectrum of 1 (Fig. 5A) exhibits a) a
resonance at 22.9 ppm attributed to the acetato methyl carbon, b) resonances in the range
48.0-60.8 ppm, attributed to methyl and methylene carbons, c) resonances in the range 114.5-
128.3 ppm and at 150.6 ppm, attributed to the presence of aromatic carbons, d) a resonance at
–
170.7 ppm in agreement with the presence of the azomethine carbon of the L1 ligand bound
to Zn(II) ion, and finally e) a resonance at 181.5 ppm attributed to the presence of the
carboxylate carbons of the acetato ligand bound to the Zn(II) ion. The 13C spectrum of 3 (Fig.
5B) exhibits a resonance at 23.2 ppm attributed to the acetato methyl carbon, b) resonances in
the range 40.0-60.2 ppm, attributed to methyl and methylene carbons, and c) resonances in the
range 117.1-121.8 ppm and resonance at 151.0 ppm attributed to the presence of aromatic
carbons. Finally, the resonance at 181.4 ppm could be attributed to the presence of the
carboxylato carbons of the acetato ligand bound to the Zn(II) ion. These observations were
subsequently confirmed by X-ray crystallography. The resonance shift was ~2.0 ppm for 1
13
and 3 downfield and was comparable to that observed in the CP-MAS C-NMR solid-state
spectrum of 1 and 3.
1
The H-NMR spectra of 1 and 3 in D2O showed several peaks. The spectrum of 1
(Supplementary Material Fig. S1A) exhibits resonances between 1.0-1.9 ppm. They are
consistent with the presence of the methyl protons (acetato groups). The resonances between
3.2-3.6 ppm are attributed to the –CH2OH group protons. The resonance at 3.7 ppm is
–
attributed to the methoxy group –OCH3 of the L1 ligand. The resonances in the range 6.6-7.1
ppm could be attributed to the presence of aromatic protons. The resonance at 8.4 ppm could
–
be ascribed to the proton of the –CH=N group of the L1 ligand bound to the Zn(II) ion. The
spectrum of 3 (Supplementary Material Fig. S1B) exhibits resonances in the range 1.1-1.8

ACCEPTED MANUSCRIPT
15
ppm. They are consistent with the presence of the methyl protons (acetato groups). The
resonance at 3.0 ppm is attributed to the –CH2OH group protons, the resonance at 3.6 ppm
could be attributed to the –N-CH2 group and the resonance at 3.7 ppm is attributed to the –
OCH3 group. The resonances in the range 6.6-7.1 ppm could be attributed to the presence of
aromatic protons. The resonance at 8.3 ppm could be ascribed to the proton of the –CH=N
–
group of the L3 ligand bound to Zn(II) ion.
3.6. Thermal Studies. The thermal decomposition of complexes 1-3 was studied by TGA
under an atmosphere of oxygen (Fig. 6). Compound 1 is thermally stable up to 174 C.
Between 174 oC and 516 oC, there is further weight loss, in line with the decomposition of the
organic structure of the molecule. No clear plateaus are reached in these stages, suggesting
that the derived products are unstable and decompose further. A plateau in the decomposition
of 1 is reached at 516 C, with no further loss up to 550 C, in line with the thesis that the
product at that temperature and beyond (516 C) is ZnO. The total weight loss of ~71.6% is
in good agreement with the theoretical value ~71.0%, in line with the following tentative
reactivity pattern:
4 Zn2(C12H16O5N)(CH3COO)3 + 79 O2
-
(L1 )
8 ZnO + 72 CO2 + 2 N2 + 50 H2O
Compound 2 is thermally stable up to 60 C. From that point on, a fairly broad heat process
points to the dehydration of 2, with the release of moisture and lattice water molecules
o
o
between 60 C and 235 C. Between 235 oC and 490 C, there is further weight loss, in line
with the decomposition of the organic structure of the molecule. No clear plateaus are
reached in these stages, suggesting that the derived products are unstable and decompose
further. A plateau in the decomposition of 2 is reached at 490 C, with no further loss up to
550 C, in line with the thesis that the product at that temperature and beyond (490 C) is
ZnO. The total weight loss of ~70.4% is in good agreement with the theoretical value
~70.2%, in line with the following tentative reactivity pattern:
4 [Zn2(C11H14O4N)(CH3COO)3].H2O + 75 O2
-
(L2 )
8 ZnO + 68 CO2 + 2 N2 + 50 H2O
o
Compound 3 is thermally stable up to 173 C. Between 173 oC and 526 C, there is further
weight loss, in line with the decomposition of the organic structure of the molecule. No clear
plateaus are reached in these stages, suggesting that the derived products are unstable and
decompose further. A plateau in the decomposition of 3 is reached at 526 C, with no further

ACCEPTED MANUSCRIPT
16
loss up to 550 C, in line with the thesis that the product at that temperature and beyond (526
C) is ZnO. The total weight loss of ~74.7% is in good agreement with the theoretical value
~74.2%, in line with the following tentative reactivity pattern:
2 Zn2(C10H12O3N)2(CH3COO)2 + 55 O2
4 ZnO + 48 CO2 + 2 N2 + 30 H2O
The described tentative assignments are consistent with previously reported results of
thermogravimetric analysis [57].
3.7. Cytotoxicity Results
3.7.1. Zn(II) toxicity in 3T3-L1 pre-adipocytes
To examine whether Zn(II) ions (non-complexed inorganic zinc) affect cell survival, 3T3-L1
pre-adipocytes were treated with 1, 10, 100 or 200 μΜ of either ZnCl2 or ZnSO4 for 24 h. Fig.
7 shows that both ZnCl2 (A) and ZnSO4 (B) affect cell survival in a concentration-dependent
manner (p<0.05). Specifically, reduction in cell survival amounts to 53 % (A) and 50% (B) at
1 μΜ, 64 % (A) and 45% (B) at 10 μΜ, and 65 % (A) and 62% (B) at 100 μΜ, and almost
100 % at 200 μΜ (p<0.05) compared to the control. Sodium deoxycholate has been used as a
positive control, exhibiting full reduction of cell survival under the employed experimental
conditions.
3.7.2. Zn(II) compound toxicity in 3T3-L1 pre-adipocytes for 24h
To assess cell toxicity of Zn(II)-complexes, 3T3-L1 pre-adipocytes were treated with various
concentrations (1-50 μΜ) of 1-3 for 24 h. Figs. 8A show that 3 is toxic even at low
concentrations (1-50 μΜ) (p<0.05), in contrast to 2, which is non-toxic compared to the
control. Toxicity for 1 starts settling in at higher concentrations (50 μΜ) with the amount of
cell death being ~33% (p<0.05). In parallel to this toxicity profiles, ligands L1H and L2H
have been found to be non-toxic in the concentration range 1-200 μΜ, with ligand L3H being
toxic throughout the same range [38].
3.7.3. Zn(II) compound toxicity in 3T3-L1 pre-adipocytes for 48h
During induction of the differentiation process, cells are exposed to Zn(II) complexes for
more than 24 h. Therefore, potential cytotoxic effects of Zn(II)-compounds were also
assessed in 3T3-L1 pre- adipocytes for 48 h. Specifically, the cytotoxic effects of 1 and 2
were investigated at various concentrations (1-50 μΜ) (p<0.05) for 48 h. Figs. 8B show that
both 1 and 2 appear to be non-toxic compared to the control for concentrations ranging from 1

ACCEPTED MANUSCRIPT
17
μΜ to 50 μΜ (p<0.05). It appears, therefore, that 1 and 2 can further be employed in the
induction of 3T3-L1 pre-adipocyte differentiation.
3.7.4. Zn(II) compound toxicity in differentiated 3T3-L1 for 48h
To examine whether the zinc compounds affect cell survival in differentiated 3T3-L1
adipocytes, cells were treated with various concentrations (1-50 μΜ) of 1 and 2 for 48 h. Fig.
9 shows that both 1 and 2 appear to be non-toxic at concentrations ranging from 1 μΜ to 50
μΜ (p<0.05). Toxicity starts settling in at concentrations beyond 50 μM (with the exception
of 1).
3.8. Cell migration results
To assess the cell migration effect of the title compounds on 3T3-L1 pre-adipocytes, a scratch
assay was performed. Cells were treated with 25 μΜ of each compound for 24 and 48h
respectively (10x 5× magnification). The migration capacity of cells grown in the presence
of only DMEM was considered as control. As shown in Fig. 10, in the case of 2 and 3, cells
exhibited decreased migration. 3 projected a complete abolishment of migration at the
concentration tested, whereas 1 appeared to have no effect on cell motility as in the case of
the control. Similar results were obtained after treatment for 48 h (data not shown).
3.9. In vitro insulin-like activity of Zn(II) compounds
In order to evaluate zinc’s insulin-like activity, the title compounds were employed to induce
cell differentiation of 3T3-L1 pre-adipocytes into mature adipocytes according to standard
protocol. Both 1 and 2 induced differentiation of pre-adipocytes into mature adipocytes, as
assessed by oil red O staining, in a concentration-dependent manner (0.5–50 μΜ).
Differentiation was evident already at 1 μM and the maximal effect was observed at 25 μΜ
(Fig. 11). This effect was comparable to the differentiation effect induced by insulin (10
ng/mL) (Fig. 11B). To further ascertain and validate the differentiation of pre-adipocytes
treated with 1 and 2, RT-PCR was run in order to determine the relative mRNA expression of
PPAR-γ, GLUT 1, GLUT 3 and GLUT 4 into mature adipocytes. An increase was observed
for PPAR–γ mRNA expression in mature adipocytes induced via differentiation by the title
compounds compared to pre-adipocytes (Fig. 12). In the case of the constitutively expressed
transporter GLUT 1, mature adipocytes differentiated with 1 and 2, showed a significant
increase in mRNA expression relative to the control, yet there was no substantial increase
compared to insulin (vide infra). In the case of GLUT 3, combinations of insulin and
compounds had a significant effect on its mRNA expression compared to the control.
However, the combination of insulin and 1 was the only one reflecting a significant increase
compared to insulin. By the same token, mature adipocytes differentiated with 1 and 2

ACCEPTED MANUSCRIPT
18
showed a 3 and 4 fold increase of GLUT 4, respectively, compared to undifferentiated pre-
adipocytes (Fig. 13). Interestingly, when 1 or 2 was used in combination with insulin in the
differentiation medium of 3T3-L1 pre-adipocytes, a higher expression of GLUT 4 and PPAR-
γ mRNA was observed in the case of 1 compared to that of insulin or 1 alone (Fig. 12 and
13C), suggesting an additive and/or synergistic effect of this compound to insulin-induced
adipogenesis.
4. Discussion
4.1. The concept and emergence of ternary Zn(II)-acetato-Schiff compounds
The coordination environment of zinc modulates profusely the biological activity of that
metal ion in the cellular milieu. In this respect, the nature and properties of the ligands-
substrates bound into the Zn(II) coordination sphere essentially formulate the potential of that
metal ion to induce phenotypic changes commensurate with the sought after effects
(adipogenicity, restoration of normoglycemia, etc.). Among those ligand properties are the
ultimate hydrophilic-hydrophobic character of the Zn(II) complex form, the ability to promote
Zn(II) chelation through (O,N)-containing terminals and thus bring about stabilization,
allowance for kinetically labile third type ligands (e.g. water) to bind the metal ion thereby
setting the stage for further biological interactions of Zn(II) with cellular targets of
importance to the reduction of hypergrlycemia, and others.
In view of the aforementioned, the substrates-ligands selected to explore the chemistry of
Zn(II) fulfill several of the stated criteria, thus principally guiding the synthesis of ternary
materials aptly poised to seek and establish interactions with molecular targets in the cellular
medium linked to the insulin-like activity of fibroblast differentiation toward adipocytes,
capable of reducing blood glucose levels in type 2 Diabetes mellitus. The vanillin-based
Schiff ligands [36-38] provide a flexible cleft for Zn(II) binding through the phenolato and
azomethine imino terminals, with the remaining available coordination sites being taken up
by either third ligands (acetates in starting reagent) and/or the alcoholic terminals in the Schiff
base. The presence of alcohols in the Schiff ligands bound to Zn(II) also provides an
opportunity for establishment of interactions (i.e. through H-bonding) with intracellular
molecular targets, thus promoting Zn(II) biological activity reflected in cytotoxicity, cell
differentiation and/or insulin mimetic activity. In such a context, L1H contains three amino
alcoholic terminals, L2H contains two and L3H contains only one. Logically, therefore, the
synthetic approaches used in this work are linked to a common source of Zn(II) a) as a
synthon containing bound acetato groups, and b) under appropriate reaction conditions
ensuring the isolation of dinuclear assemblies bearing the appropriate components (ligands

ACCEPTED MANUSCRIPT
19
and acetato groups) and carrying newly formulated chemical properties bestowed upon the
Zn(II) assemblies. The physicochemical characterization of the ligands and the associated
materials 1-3 relying on elemental analysis, FT-IR, ESI-MS, NMR and X-ray crystallography
confirm their nature, thereby justifying their further use in biological experiments.
4.2. Structure-based association with potential Zn(II) biological activity
The isolated ternary Zn(II) compounds 1-3 are chemically well-defined species, which owing
to their design and synthetic emergence are poised to seek biological activity at the molecular
level. To this end, the three materials belong to a uniform family of species, with common
features including the following: a) they are dinuclear complexes of Zn(II), b) they all contain
at least one Schiff ligand (compound 3 contains 2 ligands) bound to Zn(II), c) they all contain
acetato groups bound to Zn(II), and d) in all three compounds the Schiff ligand is singly
deprotonated, with the amino alcoholic moieties retaining their protons.
Specifically, in 1, two of the acetato ligands span over the two Zn(II) ions in the dinuclear
assembly with the third acetato ligand coordinating one of the two Zn(II) centers in a
bidentate fashion. That ligand coordination activity around the dinuclear assembly generates
an organizational asymmetry in the molecule, with the tri-alcoholic tether extending on one
end only. In this sense, it’s worth emphasizing the fact that one of the three amino alcoholic
moieties in the dinuclear assembly binds a Zn(II), while still retaining its alcoholic proton,
with the other two dangling away from the Zn(II) core. This apparent structural arrangement
might have implications on the exertion of biological behavior of the compound. In 2, the
same motif of acetato groups bound to the dinuclear assembly is recognized as in 1, with the
di-alcoholic moiety attached to the Schiff core dangling away from the dinuclear assembly
and only one alcoholic moiety remaining protonated and bound to one Zn(II) center. The so
arising asymmetry in the ligand arrangements around the dinuclear core resembles that
encountered in 1. Worth noting at this juncture is the fact that 1, containing Zn(II) bound to
–
L1 , emerges as toxic as the free ligand L1H itself, in stark contrast to the congener vanadium
compound bearing the same ligand bound to V(V). The latter species was not toxic to the
cells [38].
In 3, one of the acetato groups spanning the dinuclear Zn(II) unit has been replaced by a
second Schiff ligand bound to the unit, thereby raising the number of bound Schiff ligands to
two. The terminally bound acetato group to one of the Zn(II) centers differentiates itself from
the corresponding ones in 1 and 2 in that it binds the metal ion center in a monodentate
fashion. The so arranged ligands in the dinuclear core introduce an asymmetric character in
the molecule, reflected ostensibly in the structural features inherent to the Zn(II) core.

ACCEPTED MANUSCRIPT
20
The aforementioned description of the three molecules reveals differences among the species
that include the following: a) a variable number of coordinated acetato groups to the dinuclear
Zn(II) assemblies (three for 1 and 2 and two for 3), b) a variable coordination mode in the
terminally bound acetato groups to the Zn(II) dinuclear core, b) a variable Zn(II):Schiff base
ligand ratio (2:1 for 1 and 2 and 1:1 for 3) in the dinuclear assemblies, and c) a variable
number of amino alcoholic tethers, thereby projecting potentially differing-variable capacities
in the pursuit of interactions (i.e. hydrogen bonding) with cellular targets. Collectively, the
outlined similarities and differences in 1-3 signify some of the factors that might influence the
activity of the three species at the biological level in terms of toxicity and cellular
differentiation related to insulin mimetic activity toward fibroblast differentiation (vide infra).
In this sense, the biological profiles of 1-3 arising in the experiments could further be
amenable to optimization of the observed biological activity of Zn(II) through adjustments in
the nature and properties of the ligands seeking metal ion coordination and leading to
dinuclear Zn(II) assemblies.
4.3. Structure-selective influence of Zn(II)-compounds on pre- and mature adipocyte
survival.
Diabetes mellitus is a group of metabolic diseases mainly characterized by disregulation of
glucose metabolism. There is a significant correlation between zinc and Diabetes mellitus,
although an intricate bilateral web of complex interactions still remains unclear. Zinc can
directly affect insulin secretion, synthesis and signaling. In this respect, the present work
examined the insulin-like effect of three newly synthesized zinc compounds toward induction
of pre-adipocyte cell differentiation. The results, clearly demonstrate the differentiating
potential of zinc complexes. Zn(II) ions (from ZnCl2 or ZnSO4) appear to reduce cell
viability in a concentration-dependent manner. The cytotoxic effects of zinc ions are reduced
through complexation. Consequently, the nature of substrates to which Zn(II) is bound does
emerge crucial to the biological activity it exemplifies. In fact, the results indicate the
–
–
structure-specific influence of (L1 )-(L3 ) Zn(II)-bound ligands on pre-adipocyte survival.
Hence, the toxicity profile exemplified by the Zn(II)-Li (i=1-3) compounds mirrors that of the
free ligands. 1 and 2 appear to be non-toxic, whereas 3 influences cell survival in a
concentration-dependent manner. Interestingly, the emerging cytotoxic pattern, when pre-
adipocytes were treated with 3 for 24 h, is almost the same as that of the substrate (L3H) to
which zinc is ligated. In addition, the cytotoxic profile of each compound examined clearly
depends on the number of terminal alcohol arms. Along the same lines, a cell migration test
was performed on each zinc compound with the specific cell line. To this end, it’s worth

ACCEPTED MANUSCRIPT
21
noting the fact that the specific cell line exemplifies chemotactically its cell migration
potential. The results reveal that although a zinc compound can be non-toxic, it may have an
effect on the endogenous cell motility. The overall results indicate that 1 does not affect cell
migration following 24 h treatment as in the case of the control.
4.4. Zinc-induced adipogenesis with insulin-like activity
The adipogenic effect of insulin is one of the most fundamental functions of insulin signaling
in adipose tissue and depends on several signaling molecules, including insulin receptor
substrate 1 and 2, phosphoinositide 3-kinase, and protein kinase B. The murine 3T3-L1 cell is
an established insulin-responsive pre-adipocyte model for the study of adipogenesis. As a
result, this cell line is used in order to investigate the differentiation potential of 1 and 2. Both
compounds induced differentiation of pre-adipocytes into mature adipocytes in a
concentration-dependent manner similar to the effect induced by insulin, thereby
demonstrating the adipogenic potential of the title compounds. Increased expression of
peroxisome proliferator activated receptor gamma (PPAR-γ) and glucose transporter 4
(GLUT 4) in mature adipocytes treated with the Zn(II) compounds further confirmed the
differentiation capacity of these complexes. Increased expression of PPAR-γ a) is an early
event during differentiation of an adipocyte precursor cell toward a fully mature adipocyte,
and b) induces the expression of late-stage-differentiation genes. In turn, once a cell is
transformed into a mature adipocyte, glucose transporters belonging to the GLUT family
emerge as significant players in glucose metabolism. In this respect, the employment of three
members of that family, i.e. GLUT 1, GLUT 3 and GLUT 4, arose as a key subject of this
investigation related to biomarker-driven proof of the cell differentiation process having taken
place. To this end, in the case of GLUT 1, compounds 1 and 2 fared a) at almost the same
level of mRNA expression as insulin, and b) marginally above the untreated pre-adipocytes
(control). The combination of 1 and 2 with insulin did not result in any substantially higher
mRNA expression levels compared to the control either. The collective results a) are in line
with literature reports on GLUT 1 mRNA levels expressed at a basal level, essentially not
changing between pre- and mature adipocytes [58], and b) suggest that the title zinc
compounds 1 and 2 do not influence the expression of constitutive components (such as
GLUT 1) in glucose uptake. In GLUT 3, the pattern of mRNA expression in the case of
insulin and combinations thereof with 1 and 2 is similar to that of GLUT 4 (vide infra),
thereby indicating that the title compounds 1 and 2 can stimulate mRNA expression of
glucose transporters intimately associated with targets expressed mainly in non tissue-specific
cells (i.e. neuronal) [59,60]. The overall fold increase of the relative mRNA concentration of

ACCEPTED MANUSCRIPT
22
GLUT 1 is lower than two-fold compared to the control, thus ascertaining its constitutive role
in adipocyte physiology [61]. Concomitantly, GLUT-4 plays a major role in the metabolic
functions of adipocytes by allowing glucose transport into the cell during stimulation by
insulin. It is the main glucose transporter specific for adipose tissues and 3T3-L1 cell line
[62,63]. To this end, basal levels of GLUT 4 project a reliable biomarker for cell
adipogenesis. Glucose transport occurs relatively late in the differentiation process following
expression of PPAR-γ. Increased expression of PPAR-γ in adipose tissue results in rising
levels of GLUT-4 [64]. Inhibition of adipogenesis results in decreased expression of both
PPAR-γ and GLUT-4 [65]. The greater increase in the expression of GLUT 4 and PPAR-γ
observed upon combinational treatment of pre-adipocytes with insulin and Zn(II)-compounds
compared to insulin, and Zn(II)-compounds alone, indicates a possible additive and/or
synergistic effect between the two agents. Such insulin-like effects may involve zinc
potentiation of insulin signal transduction pathways, and/or enhancement of insulin action
through Zn(II)-compound interjection, thereby contributing to a significantly greater increase
in the expression of adipogenesis molecular factors.
5. Conclusions
The work presented here a) projects the benefits of ligand LiH (i=1-3) design exemplifying
the salient structural and electronic characteristics of ternary Zn(II)-L-acetato species 1-3 as
potential adipocyte differentiating agents mimicking insulin, b) delineates the crystallographic
and spectroscopic properties of the arising materials 1-3 through the elaborated synthetic
procedures adopted, and c) describes the biological effects brought on by the use of structure-
based selection of synthetic zinc forms in biological tests linked to insulin action on pre- and
mature adipocytes. Ligand structural differentiation exemplified through the variable number
of alcohol-containing ―arms‖-anchors in the Schiff base(s) employed, variably configuring the
nature of the arising Zn(II) species, suggests an association of the Zn(II)-bound ligand to
cellular toxicity and concomitant biological activity. The experimental account delves into
the biological effects associating Zn(II) with insulin-like adipogenesis. The collective results
reflect associations of properly configured and structurally selected ―zincoforms‖ with well-
defined biological profiles and set the stage for further physicochemically-based development
of efficient zinc antidiabetic metallodrugs.
Appendix A. Supplementary data
Supplementary NMR data on 1 and 3 in the form of figures were deposited as Supplementary
Material (Figs. S1A-B). CCDC 1059871 (1), CCDC 1059872 (2), and CCDC 1059870 (3)
contain the supplementary crystallographic data for this paper. These data can be obtained

ACCEPTED MANUSCRIPT
23
free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB21EZ, UK; fax: (+44) 1223-
336-033; or deposit@ccde.cam.ac.uk).
Abbreviation list
DM
Diabetes mellitus
IRSs
Insulin receptor substrates
Phosphatidylinositol 3-kinase
Glucose transporters
PI3K
GLUTs
PTP1B
DMSO
DMEM
PPAR-γ
EtOH
FBS
Protein tyrosine phosphatase B
Dimethyl sulfoxide
Dulbecco’s modified Eagle’s medium
Peroxisome proliferator-activated receptor gamma
Ethanol
Fetal Bovine Serum
SEM
Standard error mean
ANOVA
One way analysis of variance

ACCEPTED MANUSCRIPT
24
References
[1] J.B. Classen, Evidence childhood epidemics of type 1 and type 2 diabetes are opposite
extremes of an immune spectrum disorder induced by immune stimulants. Role of race
and associated cortisol activity as a major determining factor of the type of diabetes,
Diabetes & Metabolic Syndrome: Clinical Research & Reviews (2008)
[2] S.E. Khan, Diabetologia 46 (2003) 3–19.
[3] H. Yki-Järvinen, Diabetologia 38 (1995) 1378–88.
[4] A. Gastaldelli, E. Ferrannini, Y. Miyazaki, M. Matsuda, RA. DeFronzo, Diabetologia
47 (2004) 31-9.
[5] E. Ferrannini, A. Mari, Diabetologia 47 (2004) 943-56.
[6] L. Monnier, C. Colette, D.R. Owens, Diabetes & Metabolism 34 (2008) 207-216.
[7] H. Sakurai, A. Katoh, Y. Yoshikawa, Bull. Chem. Soc. Jpn. 79 (2006) 1645–1664.
[8] B.L. Vallee, K.H. Falchuk, Physiol. Rev. 73 (1993) 79–118.
[9] N. Wellinghausen, H. Kirchner, L. Rink, Immunol. Today 18 (1997) 519–521.
[10] A.S. Prasad, Nutrition 11 (1995) 93–9.
[11] A.B. Chausmer, J. Am. Coll. Nutr. 7 (1998) 109–115.
[12] A.Q. Truong-Tran, LH. Ho, F. Chai, PD. Zalewski, J. Nutr. 130 (2000) 1459S–1466S.
[13] A.Q. Truong-Tran, J. Carter, RE. Ruffin, PD. Zalewski, Biometals 14 (2001) 315–330.
[14] H. Haase, E. Mocchegiani, L. Rink, Biogerontology 7 (2006) 421–428.
[15] D.A. Scott, Biochem. J. 28(4) (1934) 1592–1602.
[16] S.F. Simon, C.G. Taylor, Exp. Biol. Med. 226 (2001) 43.
[17] M.M. James, S.C. Charles, J. Biol. Chem. 257 (1982) 4362.
[18] H. Sakurai, Y. Kojima, Y. Yoshikawa, K. Kawabe, H. Yasui, Coord. Chem. Rev. 26
(2002) 187.
[19] C.G. Taylor, BioMetals 18 (2005) 305.
[20] L. Coulson, P. Dandona, Diabetes 29 (1980) 665.
[21] A. Shisheva, D. Gefel, Y. Shechter, Diabetes 41 (1992) 982.
[22] M.D. Chen, S.J. Liou, P.Y. Lin, V.C. Yang, P.S. Alexander, Biol. Trace Elem. Res. 61
(1998) 303.
[23] J. Fugono, K. Fujimoto, H. Yasui, K. Kawabe, Y. Yoshikawa, Y. Kojima, H. Sakurai,
Drug Metab. Pharmacokinet. 17 (2002) 340.
[24] H. Sakurai, Y. Adachi, BioMetals 18 (2005) 319.

ACCEPTED MANUSCRIPT
25
[25] R.R. Holman, C.A. Cull, R.C. Turner, Diabetes Care 22(6) (1999) 960–964.
[26] S. Patai, The Chemistry of the Carbon-Nitrogen Double Bond, J. Wiley & Sons,
London (1970).
[27] E. Jungreis, S. Thabet, Analytical Applications of Schiff bases, Marcell Dekker, New
York (1969).
[28] D.H. Brown, W.E. Smith, Enzyme Chemistry- impact and Applications, Chapman and
Hall, London (1990).
[29] A.L. Lehninger, Principles of Biochemistry, Worth, New York (1975).
[30] M.B. Ferrari, S. Capacchi, G. Pelosi, G. Reffo, P. Tarasconi, R. Albertini, S. Pinelli, P.
Lunghi, Inorg. Chim. Acta 286 (1999) 134.
[31] E. Canpolat, M. Kaya, J. Coord. Chem. 57 (2004) 1217.
[32] M. Yildiz, B. Dulger, S.Y. Koyuncu, B.M. Yapici, J. Indian Chem. Soc. 81 (2004) 7.
[33] K. Singh, M.S. Barwa, P. Tyagi, Eur. J. Med. Chem. 42 (2007) 394.
[34] P.G. Cozzi, Chem. Soc. Rev. 33 (2004) 410.
[35] S. Chandra, J. Sangeetika, J. Indian Chem. Soc. 81 (2004) 203.
[36] M. Odabaşoğlu, C. Albayrak, O. Büyükgüngör, P. Lönnecke, Acta Cryst. C59 (2003)
o616-o619.
[37] G.X. Tana, X.C. Liub, Acta Cryst. E65 (2009) o559.
[38] E. Halevas, O. Tsave, M. P. Yavropoulou, A. Hatzidimitriou, J. G. Yovos, V.
Psycharis, C. Gabriel, A. Salifoglou, Inorg. Biochem. (2015) in press.
[39] Rigaku/MSC (2005). CrystalClear. Rigaku/MSC Inc., The Woodlands, Texas, USA.
[40] (a) Sheldrick, G.M., SHELXS-97: Structure Solving Program, University of Göttingen,
Germany, 1997. (b) Sheldrick, G.M., SHELXL-97: Crystal Structure Refinement
Program, University of Göttingen, Germany, 1997.
[41] Bruker Analytical X-ray Systems, Inc. Apex2, Version 2 User Manual, M86-E01078,
Madison, WI, 2006.
[42] Siemens Industrial Automation, Inc. SADABS: Area-Detector Absorption Correction;
Madison, WI, 1996.
[43] Palatinus, L.; Chapuis, G. J. Appl. Cryst. 40 (2007) 786-790.
[44] P.W. Betteridge, J.R. Carruthers, R.I. Cooper, K. Prout, D.J. Watkin, J. Appl. Cryst. 36
(2003) 1487.
[45] DIAMOND – Crystal and Molecular Structure Visualization, Ver. 3.1c, Crystal Impact,
Bonn, Germany, 2006.

ACCEPTED MANUSCRIPT
26
[46] L. Madsen, R. K. Petersen, M. B. Sörensen, C. Jörgensen, P. Hallenborg, L. Pridal, J.
Fleckner, E.-Z. Amri, P. Krieg, G. Furstenberger, R. K. Berge, K. Kristiansen Biochem.
J. 375 (2003) 539–549.
[47] G.B. Deacon, R.J. Phillips, Coordin. Chem. Rev. 33 (1980) 227.
[48] C.O. Rodriguez, C.O.R. de Barbarin, N.A. Bailey, D.E. Fenton, Q.Y. He, J. Chem.
Soc., Dalton Trans. (1997) 161.
[49] A. Erxleben, J. Hermann, J. Chem. Soc., Dalton Trans. (2000) 569.
[50] A. Abufarag, H. Vahrenkamp, Inorg. Chem. 34 (1995) 2207.
[51] P. Orioli, R. Cini, D. Donati, S. Mangani, J. Am. Chem. Soc. 103 (1981) 4446.
[52] M.M. Harding, S.J. Cole, Acta Crystallogr. 16 (1963) 643.
[53] R.H. Kretsinger, F.A. Cotton, R.F. Bryan, Acta Crystallogr. 16 (1963) 651.
[54] S.C. Rawle, A.J. Clarke, P. Moore, N.W. Alcock, J. Chem. Soc., Dalton Trans. (1992)
2755.
[55] A.B. Charette, J.-F. Marcoux, F. Bélanger-Gariépy, J. Am. Chem. Soc. 118 (1996)
6792.
[56] C.P. Rao, A. Sreedhara, P.V. Rao, M.B. Verghese, K. Rissanen, E. Kolehmainen, N.K.
Lokanath, M.A. Sridhar, J.S. Prasad, J. Chem. Soc., Dalton Trans., (1998) 2383-2393.
[57] C. Duval, ―Inorganic Thermogravimetric Analysis‖, Second and Revised Edition,
Elsevier Publishing Co., Amsterdam, (1963) 294-300.
[58] S. Perrini, L. Laviola, A. Cignarelli, M. Melchiorre, F. De Stefano, C. Caccioppoli, A.
Natalicchio, M. R. Orlando, G. Garruti, M. De Fazio, G. Catalano, V. Memeo, R.
Giorgino, F. Giorgino, Diabetologia 51 (2008) 155–164.
[59] T. Kayano, H. Kuomoto, T. Eddy, Y.-S. Fan, M.G. Byers, T.B. Shows, G.I. Bell,
J. Biol. Chem. 263 (1988) 12245-12248.
[60] Guillet-Deniau, A. Leturque, J. Girard, J. Cell Sci. 107 (1994) 487-496.
[61] M. Mueckler, Diabetes 39 (1990) 6-11.
[62] S.W. Cushman, L.J. Wardzala, J. Biol. Chem. 255, (1980) 4758-4762.
[63] K. Suzuki, T. Kono, Proc. Natl. Acad. Sci. U.S.A. 77 (1980) 2542-2545.
[64] E.D. Rosen, B.M. Spiegelman, J. Biol. Chem. 276 (2001) 37731–37734.
[65] T. Furuyashiki, H. Nagayasu, Y. Aoki, H. Bessho, T. Hashimoto, K. Kanazawa, H.
Ashida, Biosci. Biotechnol. Biochem. 68 (2004) 2353–2359.

ACCEPTED MANUSCRIPT
27
Table 1: Summary of Crystal, Intensity Collection and Refinement Data for
[Zn2(L1)(CH3COO)3] (1), [Zn2(L2)(CH3COO)3]∙H2O (2) and [Zn2(L3)2(CH3COO)2] (3).
1
C18H25NO11Zn2
562.13
2
C17H25NO11Zn2
550.12
3
C24H30N2O10Zn2
637.27
formula
molecular mass
T, oK
160(2)
160(2)
295(2)
radiation type
wavelength, λ (Ǻ)
crystal system
space group
a (Ǻ)
Cu Kα
Cu Kα
Mo Kα
1.54178
monoclinic
P21/n
1.54178
orthorhombic
P21cn
0.71073
monoclinic
P21/n
8.8477(2)
25.5944(5)
9.9881(2)
90
7.7448(1)
13.7545(2)
20.0501(3)
90
12.1435(11)
15.7193(16)
14.174(2)
90
b (Ǻ)
c (Ǻ)
α, deg
β, deg
109.496(1)
90
90
97.797(4)
90
γ, deg
90
V, (Ǻ3)
2132.14(8)
4
2135.85(5)
4
2680.6(5)
4
Z
Dcalcd (Mg m-3)
abs.coeff. (μ), mm-1
1.751
1.711
1.579
3.327
3.304
1.847
-10→10, -22→28,
-11→11
1.084
-9→9, -16→16,
-22→19
1.052
-15→16, -19→21,
-19→18
1.000
range of h,k,l
goodness-of-fit on F2
observed reflections
(I>2σ(I))
3471
3454
3666

ACCEPTED MANUSCRIPT
28
R (1)
0.0311(1)
0.0769(1)
0.0286(1)
0.0737(1)
0.0349(2)
0.0662(2)
(1)
Rw
(1)
2
2
R values are based on F values, Rw values are based on F2; w = 1/[σ2(Fo ) + (aP)2 + bP] where P = (Max (Fo ,0)
+2Fc2)/3
w(F 2  F 2 )2

F  F

R  
o
c
o
c
,
Rw 
w(F 2 )2



Fo

o

(2)
w′ × [1 - (ΔFobs / 6 × ΔFest)²]²
w′ =[P0T0′(x) + P1T1′(x) + ...Pn-1Tn-1′(x)]-1, where Pi are the coefficients of a Chebychev series in ti(x), and
x = Fcalc²/Fcalc max. P0 - Pn-1 =1.40 1.63 0.374
²

ACCEPTED MANUSCRIPT
29
Table 2: Bond lengths [Å] and angles [deg] for for [Zn2(L1)(CH3COO)3] (1),
[Zn2(L2)(CH3COO)3]∙H2O (2) and [Zn2(L3)2(CH3COO)2] (3).
1
2
3
Bond lengths (Å)
Zn(1)—O(1)
Zn(1)—O(2)
Zn(1)—O(11)
Zn(1)—O(13)
Zn(1)—O(15)
Zn(1)—O(16)
Zn(2)—N(1)
Zn(2)—O(2)
Zn(2)—O(3)
Zn(2)—O(12)
Zn(2)—O(14)
2.385(2)
1.970(2)
2.009(2)
2.053(2)
2.001(2)
2.497(2)
2.030(2)
2.035(2)
2.157(2)
1.955(2)
1.985(2)
Zn(1)—O(1)
1.978(2)
1.960(2)
2.043(2)
2.163(2)
2.051(2)
2.026(2)
2.041(2)
2.030(2)
2.339(2)
2.402(2)
1.987(2)
Zn(1)—O(1)
2.329(2)
2.011(2)
2.035(3)
1.984(2)
2.310(3)
2.068(3)
2.001(3)
2.065(2)
2.252(3)
1.937(2)
1.988(3)
Zn(1)—O(3)
Zn(1)—O(12)
Zn(1)—O(13)
Zn(1)—N
Zn(1)—O(2)
Zn(1)—N(2)
Zn(1)—O(4)
Zn(1)—O(5)
Zn(1)—O(7)
Zn(2)—N(1)
Zn(2)—O(2)
Zn(2)—O(3)
Zn(2)—O(8)
Zn(2)—O(9)
Zn(2)—O(2)
Zn(2)—O(4)
Zn(2)—O(5)
Zn(2)—O(6)
Zn(2)—O(11)
Zn(2)—O(12)
Bond angles (deg)
O(1)—Zn(1)—O(2)
72.86(7) O(1)—Zn(1)—O(3)
107.30(9)
95.45(9)
N(2)—Zn(1)—O(1)
N(2)—Zn(1)—O(2)
O(1)—Zn(1)—O(2)
N(2)—Zn(1)—O(4)
O(1)—Zn(1)—O(4)
O(2)—Zn(1)—O(4)
N(2)—Zn(1)—O(5)
O(1)—Zn(1)—O(5)
O(2)—Zn(1)—O(5)
93.55(11)
156.68(11)
73.58(9)
O(1)—Zn(1)—O(11) 159.08(7) O(1)—Zn(1)—O(12)
O(1)—Zn(1)—O(13)
O(1)—Zn(1)—O(15)
O(1)—Zn(1)—O(16)
O(2)—Zn(1)—O(11)
O(2)—Zn(1)—O(13)
98.22(7) O(1)—Zn(1)—O(13)
85.48(6) O(1)—Zn(1)—N
90.14(10)
133.17(9)
102.21(9)
92.08(10)
118.01(10)
91.47(11)
86.51(11)
106.70(10)
74.87(11)
80.60(11)
83.73(10)
78.37(6) O(3)—Zn(1)—O(12)
97.20(7) O(3)—Zn(1)—O(13)
98.28(7)
O(3)—Zn(1)—N
O(2)—Zn(1)—O(15) 154.30(7) O(12)—Zn(1)—O(13) 162.28(9)
O(2)—Zn(1)—O(16) 103.72(6) O(12)—Zn(1)—N 86.45(9)