Molecular recognition between adenine or 2,6-diaminopurine and copper(II) chelates with N,O2,S-tripodal tetradentate chelators having thioether or disulfide donor groups

Article abstract of DOI:10.1016/j.jinorgbio.2015.05.014

Five novel ternary copper(II) complexes with the N,O₂,S-tripodal tetradentate chelators N,N-bis(carboxymethyl)-S-benzylcysteaminate(2-) ion (BCBC) or N,N,N′,N′-tetrakis(carboxymethyl)cystaminate(4-) ion (TCC) and adenine (Hade), 2,6-diaminopuri

Full text of DOI:10.1016/j.jinorgbio.2015.05.014

Molecular recognition between adenine or 2,6-diaminopurine and copper(II)
chelates with N,O2,S-tripodal tetradentate chelators having thioether or
disulfide donor groups
Josefa Mar´ıa Gonza´lez-Pe´rez, Duane Choquesillo-Lazarte, Alicia Dom´ınguez-
Mart´ın, Hanan El Bakkali, Mar´ıa Eugenia Garc´ıa-Rubin˜o, Inmaculada Pe´rez-
Toro, Ester V´ılchez-Rodr´ıguez, Alfonso Castin˜eiras, Valeria Marina Nurchi,
Juan Niclo´s-Gutie´rrez
PII:
DOI:
Reference:
S0162-0134(15)30005-2
doi: 10.1016/j.jinorgbio.2015.05.014
JIB 9730
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
22 January 2015
26 May 2015
29 May 2015
Please cite this article as: Josefa Mar´ıa Gonz´alez-P´erez, Duane Choquesillo-Lazarte,
Alicia Dom´ınguez-Mart´ın, Hanan El Bakkali, Mar´ıa Eugenia Garc´ıa-Rubin˜o, In-
maculada P´erez-Toro, Ester V´ılchez-Rodr´ıguez, Alfonso Castin˜eiras, Valeria Ma-
rina Nurchi, Juan Nicl´os-Guti´errez, Molecular recognition between adenine or 2,6-
diaminopurine and copper(II) chelates with N,O2,S-tripodal tetradentate chelators hav-
ing thioether or disulfide donor groups, Journal of Inorganic Biochemistry (2015), doi:
10.1016/j.jinorgbio.2015.05.014
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
Molecular recognition between adenine or 2,6-diaminopurine and copper(II) chelates with
N,O2,S-tripodal tetradentate chelators having thioether or disulfide donor groups
b
a,§
Josefa María González-Pérez a,*, Duane Choquesillo-Lazarte , Alicia Domínguez-Martín
,
a
c
a
Hanan El Bakkali , María Eugenia García-Rubiño , Inmaculada Pérez-Toro , Ester Vílchez-
Rodríguez a, Alfonso Castiñeiras d, Valeria Marina Nurchi e, Juan Niclós-Gutiérrez a
a
Department of Inorganic Chemistry, Faculty of Pharmacy, University of Granada, 18071
Granada, Spain
b
, IACT, CSIC-University of Granada, Avenida de las
Palmeras 4, 18100 Armilla (Granada), Spain
c
Department of Organic and Pharmaceutical Chemistry, Faculty of Pharmacy, University of
Granada, 18071 Granada, Spain
d
Department of Inorganic Chemistry, Faculty of Pharmacy, University of Santiago de
Compostela, 15782 Santiago de Compostela, Spain
e
Dipartimento di Scienze Chimiche e Geologiche, Università di Cagliari, Cittadela
Universitaria, 09042 Monserrato, (CA), Italy
§ Present address:
Department of Chemistry, University of Zurich, CH-8057 Zurich, Switzerland
Phone: +41 446354640, E-mail: alicia.dominguez@chem.uzh.ch
* Corresponding author. Tel: +34 958243855; fax: +34 958246219.
E-mail address: jmgp@ugr.es (J.M. González-Pérez).

ACCEPTED MANUSCRIPT
Abstract
Five novel ternary copper(II) complexes with the N,O2,S-tripodal tetradentate chelators N,N-
bis(carboxymethyl)-S-benzylcysteaminate(2-) ion (B B ) N,N,N’,N’-
tetrakis(carboxymethyl)cystaminate(4-) ion (TCC) and adenine (Hade), 2,6-diaminopurine
(H p), 2,2’-bipyridine (bpy) or 1,10-phenanthroline (phen) as co-ligands were synthesized and
characterized by X-ray diffraction and other physical methods: [Cu2(BCBC)2(µ2-N3,N7-
H(N9)ade)(H2O)2]·H2O (1), [Cu2(BCBC)2(µ2-N7,N9-H(N3)dap)(H2O)2]·4H2O (2), [Cu2(µ2-
TCC)(H(N9)ade)2(H2O)2]·10H2O (3), [Cu2(µ2-TCC)(bpy)2]·15H2O (4) and [Cu2(µ2-
TCC)(phen)2]·14H2O (5). The crystal structure of H4TCC·3H2O was also determined. All
ternary Cu(II) complexes have molecular structures. The N-(2-mercaptoethyl)-iminodiacetate
moieties of BCBC or TCC ligands play a NO2+S-tripodal tetradentate role, with the S-(thioether
or disulfide) atom as the apical/distal donor of the copper(II) center. In 1-3, the iminodiacetate
moiety exhibits a mer-NO2 conformation (two nearly coplanar chelate rings) while in 4 and 5
(with bpy or phen as coligand) it displays a fac-NO+O(apical/distal) conformation. We
conclude that the formation of the Cu-S(thioether or disulfide) bonds is strongly favored by the
N-branched topology of the S-ligands in the reported compounds.
Keywords
Copper(II) complexes, Thioether, Disulfide, Interligand interactions, Adenine
2

ACCEPTED MANUSCRIPT
1. Introduction
Metal complexes with bio-ligands or synthetic pharmaceuticals containing soft sulfur donor
atoms have focused a great interest over past decades. It is currently understood that the
remarkably soft, negatively charged, thiolate atom (i.e. that in the side-chain of cysteinate
residues) is a good sigma-donor albeit poor pi-acceptor atom. Thiolate plays important roles in
biological systems [1] such as (a) approaching the relative stabilities of copper oxidation states,
matching the Cu(I)/Cu(II) couple in blue copper proteins, as well as the blue copper centers of
other metallo-proteins, (b) metal binding of Zn(II), Cu(I) and Cd(II) ions in metallothioneins,
(c) building the stable Fe-S cluster centers in iron-sulfur proteins or (d) scavenging soft metal
ions from biological systems. In contrast, neutral S-thioether (also involved in red-ox metal sites
of blue copper proteins, i.e. that in methionine side-chain) is considered a poorer sigma-donor
than pi-acceptor atom. A similar behavior could be expected for the S-atoms of disulfide
groups. It is assumed that the metal binding affinity follows the general order S-thiolate > S-
thioether > S-disulfide. Consequently, the thiolate and to a lesser extent the S-thioether atoms
play relevant metal binding roles in biological and bio-inspired model systems [2,3].
The overlap of red-ox and coordination processes related to the interaction of Cu(II) ions and
thiols, that can yield soft copper(I) atoms and the corresponding oxidized disulfide forms, is
well documented [4]. Indeed mixed-valence copper cores have been structurally characterized
for the systems Cu(I,II)-D-penicillaminate-Cl [5] and Cu(I,II)-β,β-dimethylcysteaminate [6],
although their significance within the D-penicillamine treatment of Wilson disease is still
uncertain. In contrast, there is a rather limited evidence of metal ion interactions with disulfide
groups present in bio-ligands. Coordinating reactions between disulfide groups (as oxidized
forms of thiols) and Cu(II) ions are not altered by red-ox chemistry and can yield complexes
with more or less strong CuII-S(disulfide) bonds. In this sense, the structures of Cu(II)
complexes with D-penicillamine disulfide (D-pds) [7] or oxidized glutathione (GSSG) [8]
revealed that the disulfide group can act as monodentate as well as bridging-bidentate for
copper(II) centers, respectively.
3

ACCEPTED MANUSCRIPT
This work aims to deepen on the knowledge of ternary copper(II) complexes having a chelator
with S-thioether or S-disulfide group and a purine-like ligand. To this purpose, the N,N-
bis(carboxymethyl)-S-benzylcysteaminate(2-)
(BCBC)
n
N,N,N’,N’-
tetrakis(carboxymethyl)cystaminate(4-) (TCC) ions and adenine (Hade) or 2,6-diaminopurine
(Hdap) have been selected as chelators and N-coligands, respectively (Fig. 1). In the Cambridge
Structural Database (CSD, version 5.34, updated May 2014) there are not structural evidences
for the free H2BCBC acid or BCBC-metal complexes. The molecular and/or crystal structure of
the binary Cu(II)-TCC chelate (a polymer with formula {[Cu2(µ2-TCC)(H2O)2]·4H2O}n [9]),
and two mixed-ligand complexes [Cu2(µ2-TCC)(Him)2(H2O)2]·2H2O [10] and [Cu2(µ2-
TCC)(H5Meim)2]·2H2O [11] (Him = imidazole, H5Meim = h ‘ m ’ m 5-
methylimidazole of H(4/5)Meim) have been previously reported by our research group. In all
these compounds, the disulfide group of TCC acts as binding-bidentate group between two
Cu(II) centers where each half of TCC plays a NO2S-tripodal tetradentate chelating role. The
crystallographic study of the novel compounds represents our major goal. This high resolution
technique proves the mixed-ligand nature of the novel compounds and also allows describing in
detail the metal-binding pattern of the chelators to the N-coligands. In this context, Sóvágó et al.
concluded from solution data that glycyl-L-cysteine disulfide, (HGly-Cys)2, was able to bind
Ni(II), Cu(II) and Zn(II) giving complexes without metal-S(disulfide) bonds, which was
supported by the crystal structure determination of the {[Ni(Gly-Cys)2]·H2O}n complex [12]. A
recent study by Chakravarty et al. [13] on ternary Cu(II) complexes derived of S-methyl-L-
cysteinate(1-) n (SM ) n N,N’- n α,α’-aromatic diimines (phenantholine and
closely related aromatic diimines) also revealed that the Cu-S(thioether) bond is missing in
complexes with general formula [Cu(SMC)(phen or related N-coligand)(H2O)].
< Insert Figure 1 >
4

ACCEPTED MANUSCRIPT
2. Materials and methods
2.1. Reagents
All used reagents (except for the H2BCBC·0.5H2O and H4TCC·3H2O acids) and solvents are
commercially available in Sigma-Aldrich, Acros or other suppliers, and were used as received.
2.2. Synthesis of S-containing chelating ligands and metal complexes
Syntheses for the S-containing ligands used in the present study have already been reported by
Chiotellis et al. [2] and others. The here reported procedures are optimized methods that yield
defined hydrated samples of high purity. Indeed, they have been successfully used for the
synthesis of ternary metal complexes. In addition, suitable X-ray single-crystals of one of the
ligands have also been obtained by first time (see 2.2.2 and 3.1).
2.2.1. N,N-bis(carboxymethyl)-S-benzylcysteamine acid hemihydrate (H2BCBC·0.5 H2O)
A cooled aqueous solution of KOH (30.9 g, 0.55 mol in 200 ml of water) was drop-wise added
under N2 flow to an aqueous solution of chloroacetic acid (18.9 g, 0.20 mol in 100 ml) placed in
a three-neck round-bottom flask, maintaining the reaction mixture under 0 ºC with the aid of an
ice-salt bath. To the resulting alkaline solution of potassium chloroacetate, an aqueous solution
of S-benzylcysteamine hydrochloride (20.4 g, 0.10 mol in 100 ml) was slowly added. The
solution was then heated to 80 ºC and stirred under N2 flow for 5 hours. After cooling at room
temperature, the reaction mixture was acidified to pH ~6 with 6 M HCl and then concentrated at
reduced pressure to a half of its initial volume. Afterwards, the solution was placed in an ice-
NaCl bath to force the precipitation of a white product. This solid was identified as KCl (the
main by-product) and removed by filtration. The partially desalinized mother liquors were
diluted into 100 ml of water and again acidified (pH ~2.5) with 6 M HCl. The solution was then
placed once more in an ice-salt bath to accomplish the precipitation of a crude white powder
(~20 g). This powder was dissolved in water (200 ml) with the aid of stirring and the addition of
5

ACCEPTED MANUSCRIPT
small amounts of NaHCO3 (from a weighted sample in excess). The further addition of HCl 6 M
(equivalent to the required amount of NaHCO3) produced the precipitation of the final product,
a white sample of H2BCBC·0.5H2O acid. The acid was collected by filtration, washed with cool
water, ethanol and diethyl ether and then air dried. Yield: 14.25 g (48.7 %). Elemental analysis,
calculated for C13H18NO4.5S: C 53.41, H 6.21, N 4.79, S 10.97. Experimental: C 53.78, H 6.25,
N 4.82, S 10.77. 1H NMR (300 MHz, DMSO-d6, SD2.1.b) δ 7.52 ( , J = 4.0 Hz, 4H), 7.46 (m,
1H), 3.94 (s, 2H), 3.63 (s, 4H), 3.03 (m, 2H), 2.70 (m, 4H). 13C NMR (75 MHz, DMSO-d6,
SD2.1.b) δ 172.44, 138.84, 128.81, 128.32, 126.73, 54.53, 53.66, 35.01, 28.79. FT-IR (cm-1,
KBr disk, SD2.2.b): νas(H2O)+ν(O-H) of - OOH 3425, νs(H2O) ~3250, ν( -H)arom 3027,
νas(CH2) 2971 n 2924, νs(CH2) 2862. A broad (due to H-bonding interactions) and intense
band at 1692 v p h n n h m ν( =O) – OOH, δ(H2O), νas(COO) of –
COO– n δ(N+-H), ν( -O)+δ(O-H) of –COOH 1401. νs( OO) 1359, δ(O-H) 1281 n π(O-H)
876 of – OOH, π( -H)arom 776 and 709 (five adjacent aromatic C-H bonds usually show these
bands at 770-730 and 710-690), ν( -S) 615.
2.2.2. N,N,N’,N’’-tetrakis(carboxymethyl)cystamine acid three-hydrate (H4TCC·3H2O)
This synthesis was performed in a similar way as described in the previous section. A solution
of chloroacetic acid (23.6 g, 0.25 mol in 100 ml of water) was cooled (< 0 ºC) in a water-salt
bath and then mixed with a cool solution of KOH (28.1 g, 0.50 mol in 100 ml of water) under
N2 flow and stirring. An alkaline aqueous solution was prepared by adding small portions of
cystamine dihydrochloride (10.2 g, 0.05 mol) on a cooled KOH solution (5.6 g, 0.1 mol in 50 ml
of water). This cystamine solution was drop-wise added to the alkaline solution of potassium
chloroacetate and the resulting reacting mixture was left to stand at room temperature in the
dark during 2-3 weeks. The solution was then filtered under vacuum to remove undissolved
materials and acidified to pH 2.5 with HCl 6 M. This acid solution was concentrated under
reduced pressure at 40-45 ºC to a third of the initial volume and again cooled down in an ice-salt
6

ACCEPTED MANUSCRIPT
bath in order to induce the precipitation of KCl (main by-product), which was then removed by
filtration. The clear mother liquors were kept at 5 ºC during some weeks in a round bottom
flask, scratching its walls with a glass rod every day to induce the precipitation of the final
product. Finally, a white powder that was collected by filtration, washed with water and acetone
and then air dried. This material was re-crystallized in water to yield samples of H4TCC·3H2O.
A typical overall yield is ~50-55 %. Caution! In this synthesis, the use of a significant excess of
potassium chloroacetate (larger than 25 %) breaks the disulfide bond of cysteamine and as a
result the synthesis affords the N,N,S-tris(carboxymethyl)cysteamine acid [HO2CCH2-S-
C2H4N(CH2CO2H)2] instead of the desired product. E.A., calculated for C12H26N2O11S2: C
32.67, H 6.08, N 6.33, S 14.59. Experimental (averaged values): C 32.87, H 5.98, N 6.39, S
14.63. 1H NMR (300 MHz, DMSO-d6, SD2.1.a) δ 6.57 (m, 4H), 3.54 ( , 4H), 3.03 ( , J = 8.3,
5.5 Hz, 2H), 2.89 (dd, J = 8.4, 5.5 Hz, 2H). 13C NMR (75 MHz, DMSO-d6, SD2.1.a) δ 172.81,
104.68, 55.00, 53.90. FT-IR (cm-1; KBr disk, S.D.2.2.a): νas(H2O)+ν(O-H) of -COOH 3447,
νs(H2O) ~ 3265 (sho ), ν( -H)arom 3021, νas(CH2) 2972 n 2920, νs(CH2) 2832, ν( =O)
–COOH 1722 and 1688 (H-bonded to N+-H and O-H n p ), δ(H2O) 1627, νas(COO) of
–COO– v y m h h ~1590, w k n δ(N+-H) is expected at 1540-1500 but is
not observed, ν( -O)+δ(O-H) of – OOH 1401, νs( OO) 1389, δ(O-H) 1250 n π(O-H) 942 of
–COOH, and ν(S-S) 532 (weak). Plated colorless single-crystals were ground by slow diffusion
of diethy ether into a saturated aqueous solution of this acid.
2.2.3. Synthesis of [Cu2(BCBC)2(µ2-N3,N7-H(N9)ade)(H2O)2]·H2O (1)
A reaction with equimolar amounts of Cu(II), BCBC and Hade yielded a product which
elemental analysis clearly indicates a stoichiometric ratio 2:2:1. In addition, isolated white
crystals were also collected from the mother solutions which were identified by FT/IR
spectroscopy as free adenine. Hence, the synthesis of 1 was re-designed as follows. Bluish basic
copper(II) carbonate (Cu2CO3(OH)2, 0.5 mmol, 0.11 g) and the acid H2BCBC·0.5H2O (1 mmol,
7

ACCEPTED MANUSCRIPT
0.29 g) were reacted in 100 ml of water with stirring, heating (~50 ºC) and vacuum for half an
hour. Heating and stirring was maintained for one more hour. The blue solution was cooled to
room temperature and then filtered without vacuum. Adenine (0.5 mmol, 0.14 g) was added to
the previous solution with continuous stirring and kept for half an hour until obtaining a clear
solution. This reacting mixture was then filtered in a crystallizing dish, which was covered with
a plastic film to avoid contamination and control the evaporation of the solvent. In a few days,
blue needle-like crystals appeared (SD4.1). Over some weeks, samples of the desired product
were collected by filtration, with a yield of 78 % (0.34 g). Suitable crystals for X-ray diffraction
purposes were obtained along all crystallization process. Elemental analysis, calculated for
C31H41Cu2N7O11S2: C 42.36, H 4.70, N 11.16, S 7.30. Experimental C 42.18, H 4.75, N 11.07, S
7.21. FT-IR (cm-1; KBr disk, SD4.2.a): νas(H2O) 3430, νas(NH2) 3337 , νs(H2O)+νas(NH2) ~
3250 (smooth shoulder), ν(N-H) 3145, ν( -H)arom 3031, νas(CH2) 2967 and 2922, νs(CH2) 2861,
intense absorption at 1600-1700 nv v h n n m δ(H2O) at ~1636, νas(COO)
n δ(NH2) 1612 n δ(N-H) 1526, νs( OO) 1385, π( -H)arom of BCBC 772 and 705 (five
adjacent aromatic C-H bonds give sharp peaks at 770-730 and 710-690), ν( -S) 634. Electronic
(diffuse reflectance) spectrum (SD4.2.b): d-d band w h νmax at 14200 cm-1 and shoulder 10000
cm-1. Despite the reduced hydration of the compound, under air-dry flow it losses non-
coordinated water leading to a partially dehydrated sample with averaged formula
[Cu2(BCBC)(µ2-N3,N7-H(N9)ade)(H2O)2]·0.4H2O. All remaining water is lost between 55-115
ºC (for n = 2.4 H2O, % Calc. 4.981; Found 4.924). The thermal decomposition (SD4.3) of the
organic ligands produces H2O, CO2, CO, SO2, N2O, NO, NO2 and probably SCNH acid. The
weight is stabilized at 570 ºC (20.019 %). Additional weight loss occurs > 800 ºC yielding an
experimental residue of 18.881 % at 950 ºC (calculated for CuO 18.326%).
2.2.4. Synthesis of [Cu2(BCBC)2( µ2-N7,N9-H(N3)dap)(H2O)2]·4H2O (2)
8

ACCEPTED MANUSCRIPT
A synthetic procedure similar to that described for compound 1 was followed but using 2,6-
diaminopurine (1 mmol, 0.15 g) instead of Hade. Yield was ca. 75 % (0.36 g). Elemental
analysis calculated for C31H48Cu2N8O14S2: C 39.28, H 5.10, N 11.82, S 6.77. Experimental C
39.12, H 5.28, N 11.82, S 6.71. FT-IR (cm-1; KBr k): νas(H2O) 3406, νas(NH2) 3337,
νs(H2O)+νas(NH2) ~ 3250, ν(N-H) 3151, ν( -H)arom 2983, νas(CH2) 2952 n 2928, νs(CH2)
2859. An intense absorption near 1650 involves h n n δ(H2O) 1636, νas(COO)
n δ(NH2). A w k p k δ(N-H) at 1522, νs( OO) 1382, π( -H)arom of BCBC at 772 and
706, π( -H)arom H p v y w k p k 859, ν( -S) 654.
2.2.5. Synthesis of [Cu2(µ2-TCC)(H(N9)ade)2(H2O)2]·10 H2O (3).
Small amounts of the binary chelate [Cu2(µ2-TCC)(H2O)2]·6H2O were prepared by reaction of
metallic Cu (in excess) and the H4TCC acid in water exposed to the oxygen of the air [9]. The
reaction initially produced a violet Cu(I)-thiolate intermediate. In a test tube, this violet solution
was slowly oxidized by O2 (in contact to the air) to yield a greenish solution where well-shaped
single crystals were formed, some of them suitable for X-ray diffraction purposes [9]. Larger
amounts of the binary compound (in solution), to be used as starting reactant in the synthesis of
mixed-ligand metal complexes, were also prepared by equimolar reaction of Cu2CO3(OH)2 and
H4TCC in water [10,11]. Compound 3 was prepared by the drop-wise addition of an aqueous
solution of Hade (0.36 mmol, 48 mg, 30 ml) to an aqueous solution of the binary chelate
[Cu2(µ2-TCC)(H2O)2]·6H2O (0.18 mmol, 111 mg) in 100 ml of water. The reaction mixture was
filtered without vacuum on a crystallizing dish and covered with a plastic film. Slow
evaporation of the solvent yielded well-shaped parallelepiped single crystals (SD5.1). Yield is
~70 %. Freshly prepared samples repeatedly agreed to the formula [Cu2(µ2-
TCC)(Hade)2(H2O)2]·10H2O although the deca-hydrate seems not to be very stable. Elemental
analysis calculated for the deca-hydrate: Calculated for C22H50Cu2N12O20S2: C 26.59, H 5.07, N
16.91, S 6.45. Experimental C 26.87, H 4.78, N 17.08, S 6.63. FT-IR (cm-1; KBr disk, SD5.2.a):
9

ACCEPTED MANUSCRIPT
νas(H2O)+νas(NH2) 3410, νs(H2O)+νas(NH2) ~ 3230 , ν( -H)arom n n , νas(CH2) 2979 and
2924, νs(CH2) 2862, δ(H2O)+ δ(NH2)+νas(COO) and skeletal Hade absorptions overlapped in a
n ~1625, δ(N-H) 1525 (w k), νs( OO) 1383, π( -H)of Hade n n , ν( -
S) 637, ν(S-S) 532 (very weak). Electronic (diffuse reflectance) spectrum: asymmetric d-d band
with In air dry flow the compound losses water to give a tetra-hydrated sample according with
νmax 14.410 cm-1 (intensity barycenter at 14.200 cm-1) and smooth shoulder near 10.000 cm-1
(SD5.2.b). In air dry flow the compound losses water to give a tetra-hydrated sample according
to the averaged TG-formula [Cu2(µ2-TCC)(H(N9)ade)2(H2O)2]·4H2O (SD5.3). This sample
losses the rest of solvent molecules and the aqua ligands in overlapped steps (25-150 ºC, exp.
12.272 %, calc. 12.202 %) and then decomposes the organic ligands producing H2O, CO2, CO,
C4H4, SO2, N2O, NO and NO2, to yield a stable residue of 24.976 % at 600 ºC. It further
decomposes (at 670-730 ºC) to probably yield a non-pure CuO residue (exp. 20.933 %, calc.
17.959 %).
2.2.6. Synthesis of [Cu2(µ2-TCC)(bpy)2]·15H2O (4)
Cu2CO3(OH)2 (0.5 mmol, 0.11g) and H4TCC·4H2O (0.5 mmol, 0.22 g) were reacted in a
Kitasato flask with moderate heating (50 ºC), stirring and vacuum during one hour. The heating
and stirring (without vacuum) continued for 24 hours. Filtration was sometimes needed to
obtain a clear blue solution of the binary chelate. To this solution, 2,2’-bipyridine (1 mmol,
0.16g) was added. The solution was stirred at room temperature until complete solution of the
aromatic α,α’-diimine and then filtered into a crystallizing dish which was covered with a
plastic film and left to stand at room temperature. After two months, dark blue parallelepiped
crystals of the ternary compound were formed, many of them suitable for X-ray purposes. These
crystals were unstable out of the solution and were easily re-dissolved at room temperature.
However, well-shaped crystals newly appeared if the saturated solution was left to stand in a
stopped Eppendorf flask at 5º C. The yield was estimated as > 70 %. Analytical data of freshly
10

ACCEPTED MANUSCRIPT
prepared samples do not agree with a specific hydrated formula. Hence, the elemental
composition was determined by the crystallographic study of a crystal picked off from the
solution, appropriately protected, which affords the formula for the pentadeca-hydrate
C32H62Cu2N6O23S2 . FT-IR (cm-1; KB k): νas(H2O) 3426 , νs(H2O) ~ 3250 ( h ), ν( -
H)arom 3033, νas(CH2) 2963 n 2929, νs(CH2) ~2850, δ(H2O) 1637 n / 1616, νas(COO) 1590,
νs( OO) 1385, π( -H) 773 and 733 (four adjacent aromatic C-H bonds give a sharp band to this
deformation mode at 770-735), ν( -S) 638 n ν(S-S) 532 (very weak).
2.2.7. Synthesis of [Cu2(µ2-TCC)(phen)2]·14H2O (5)
This compound was prepared as noted above for compound 4 but using 1,10-phenantroline (0.1
mmol, 0.18 g) instead of bpy. Again, the crystals were highly hydrated and spontaneously lose
water out of the mother liquors or were re-dissolved at room temperature. Cooling the saturated
solution at 5 ºC led to nice crystals. A yield of > 70 % is easily obtained. Analytical data of
freshly prepared samples do not agree with a specific hydrated formula. Hence, the elemental
composition was determined by the crystallographic study of a single-crystal picked off from
the mother liquors, appropriately protected. It affords the tetradeca-hydrate C36H60Cu2N6O22S2.
FT-IR (cm-1; KB k): νas(H2O) 3420 , νs(H2O) ~ 3250 ( h ), ν( -H)arom 3064, νas(CH2)
2972 n 2921, νs(CH2) ~2850, δ(H2O) 1630, νas( OO) 1596, νs( OO) 1391, π( -H) 853 (two
adjacent aromatic C-H bonds give a sharp band 860-800) and 812 (three adjacent aromatic C-H
bonds give a sharp band 810-750), ν( -S) 647, ν(S-S) was not identified.
2.3. X-ray crystallography
Suitable crystals were mounted on glass fibers and these samples were used for data collection.
Data were collected with the diffractometers Bruker D8 Venture (H4TCC·3H2O), Bruker
SMART 1000 (compounds 1, 2 and 5), or Bruker X8 Kappa APEXII (compounds 3 and 4). The
data were processed with SAINT (1, 2 and 5) [14] or APEX2 (H4TCC·3H2O, 3 and 4) [15] and
11

ACCEPTED MANUSCRIPT
corrected for absorption using SADABS [16]. The structures were solved by direct methods
[17], which revealed the position of all non-hydrogen atoms. These atoms were refined on F2 by
a full-matrix least-squares procedure using anisotropic displacement parameters [16]. All
hydrogen atoms were located in difference Fourier maps and included as fixed contributions
riding on attached atoms with isotropic thermal displacement parameters 1.2 times those of the
respective atom. H-atoms of non- coordinated water molecule in 1 were not located. In 3,
solvent molecules (observed from the difference Fourier maps) could not be identified because
of significant disorder and/or partial occupancy, they were SQUEEZE'd out using PLATON
[18]. Geometrical calculations were carried out and drawings were produced with PLATON
[18] and MERCURY [19]. A summary of crystal data is shown in Table 1. Crystallographic
data for the structural analysis have been deposited at the Cambridge Crystallographic Data
Center, No. CCDC 1043608 for H4TCC·3H2O (SD1) and Nos. 1043346 -1043350 for
copper(II) derivatives 1-5 (SD3.1-SD3.5) respectively. Copies of this information may be
obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
(fax: 44 1223 336 033; e-mail: deposit@ccdc.cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
2.4. Other physical methods
Analytical data were obtained in a Thermo-Scientific (Flash 2000) elemental micro-analyzer.
FT-IR spectra were recorded using KBr pellets on a Jasco FT-IR 410 or a Jasco FT-IR 6300
spectrometer. Thermogravimetric (TG) analyses were carried out in air-dry flow (100 ml/min)
by a Shimadzu thermo-balance TGA-DTG-50H instrument, coupled with a FT-IR Nicolet
Magma 550 spectrometer. A series of FT-IR spectra (20-30 per sample) of the evolved gasses
were time-spaced recorded during the TG experiment. Diffuse reflectance (electronic) spectra
were recorded in a Varian Cary-5E spectrophotometer.
3. Results and discussion
12

ACCEPTED MANUSCRIPT
3.1 S N,N,N’,N’’-tetrakis(carboxymethyl)cystamine acid three-hydrate
The asymmetric unit of H4TCC·3H2O contains one molecule of the disulfide acid and three
crystallization water molecules. The crystal structure is stabilized by classical N-HO and O-
HO hydrogen bonds and weak C-HO hydrogen bonds (SD1), generating a three-dimensional
network.
According to its diamino-tetracarboxylic formula, the H4TCC molecule adopts a double
zwiterionic structure. In the crystal, each H4TCC molecule is attached to the two nearest
neighbors through inter-molecular hydrogen bonds O21-H21O13#2 (#2: 1/2+x, 1/2-y, 1/2+z;
2.446(3) Å, 172.9º) and O21-H21O14#2 (3.183(3) Å, 114.4º), resulting in an infinite 1D-
chain parallel to the a x (F . 2). Th ‘ n m’ N1-H1 bond is also involved in the inter-
molecular interaction N1-H1···O2(water) [2.783(3) Å, 143.8º]. The intra-molecular H-bonds
are N1-H1···O14(carboxylate) [2.682(3) Å, 98.7º] and N1-H1···O12(carboxylic) [2.729(3) Å,
108.2º]. Hence, half of the H4TCC molecule has its H1-onium proton located in a distorted
surrounding [N1, O2, O12, O14]. In contrast, the other half-H4TCC unit has the N2-H2 ‘ n m’
moiety involved in the inter-molecular interaction N2-H2···O24#3(carboxylate) (#3: x+1, y, z;
2.748(3) Å, 158.0º). One additional intra-molecular H-bonding interaction can be considered
nv v n h H2 ‘ n m’ m y [N2-H2···O22(carboxylic) (2.706(3) Å, 101.9º)]. Therefore,
the onium-H2 atom in the second half-H4TCC is located in a distorted trigonal surrounding [N2,
O22, O24]. In this compound, the S-S bond distance is 2.9494(1) Å and the C-S-S-C torsion
angle value is -95.65º.
< Insert Figure 2 >
3.2. Ternary copper(II) complexes with the tripodal tretradentate chelator BCBC.
13

ACCEPTED MANUSCRIPT
In the CSD database, there are no previous structural records regarding the H2BCBC acid or its
metal complexes. Any attempt to isolate suitable crystals for X-ray diffraction of the binary
copper(II) chelate was unsuccessful.
3.2.1. Ternary complex with BCBC and bridging Hade as coligand.
The structure of compound [Cu2(BCBC)2(µ2-N3,N7-H(N9)ade)(H2O)2]·H2O (1) is shown in
Fig. 3. The asymmetric unit consists of a binuclear molecular complex and a non-coordinated
water molecule. In the complex molecule (Fig. 3, top) both metallic centers are achiral and have
the stereochemical descriptor OC-6-32. The two non-equivalent metallic centers show a similar
elongated octahedral surrounding, type 4+1+1. The trans-apical coordination sites are occupied
by the O-aqua donor and the S-thioether donor from the tripodal tetradentate BCBC chelator.
The equatorial coordination plane of each metal center comprises the N(amino) and two
O(carboxylate) donors from the iminodiacetate (IDA) moiety of an BCBC ligand plus one N-
donor atom (N3 or N7) from the Hade nucleobase. Thus, the IDA moiety of BCBC adopts a
mer- conformation (two metal-glycinate rings being nearly coplanar and sharing the
corresponding Cu-N(BCBC) bond). Adenine exhibits its most stable tautomer [H(N9)ade] and
the μ2-N3,N7 bridging bidentate role. The coordination bonds of the equatorial plane show
distances close to 2 Å, while the Cu-(O)aqua bond distances are about 2.4 Å (see SD3.1). In this
molecule, the Cu-S(thioether) bonds (Cu1-S1 2.812(3) Å and Cu2-S2 2.977(3) Å) exhibit rather
usual bond distances as compared to many other Cu(II) complexes with the S-thioether donor in
an apical position of a copper(II) surrounding [2]. Note that these bonds are significantly shorter
than the sum of the corresponding Van der Walls radii of the implied atoms (1.40 (Cu) + 1.80
(S) = 3.2 Å). In this molecule, both the Cu1-N3 and Cu2-N7 coordination bonds cooperate with
the corresponding intra-molecular interligand H-bonding interactions (N9-H···O(coord.)
(2.698(9) Å, 140.0º) and N6-H···O(coord) (2.782(9) Å, 174.7º), respectively). This metal
binding pattern has been previously described for the most stable neutral adenine tautomer
14

ACCEPTED MANUSCRIPT
H(N9)ade in the tetranuclear compound [Cu4(pheida)2(µ2-pheida)(µ2-Hade)2(H2O)2)]·4H2O
(pheida = N-phenethylimionodiacetate(2-) ion) [20] and the
dinuclear
complex
[Cu2(pdc)2(H2O)2(2-Hade)]·3H2O [21] (pdc = pyridine-2,6-dicarboxilate(2-) ion). For the rigid
pdc ligand the corresponding mononuclear complex [Cu(pdc)(H(N9)ade)] has been previously
reported [21]. In this case, the Cu-N7(Hade) bond is also reinforced by an N6-
H···O(coordinated carboxylate) interligand interaction [21]. In contrast, neither Cu(BCBC) nor
Cu(pheida) chelates have proved to be able to yield the corresponding mononuclear complexes
with general formula Cu(chelator)(Hade) [20]. Note that both pheida and BCBC chelators have
large and flexible N-aryl-alkyl (pheida) or N-aryl-pseudoaliphatic (BCBC) pendant arms linked
to the iminodiacetate moiety. In compound 1, the strategic position of the S-thioether atom in
the BCBC skeleton favors the copper(II) coordination as a part of the NO2S-tripodal tetradentate
role. This fact restricts the conformational freedom of the N-(aryl-peudo-alkyl) moiety in the
BCBC pendant arm and might favors the crystallization of 1 (and also that of compound 2- see
below).
In the crystal of 1 no π,π,-stacking interactions are observed between the aromatic moieties of
the ligands. Double-stranded chains of complex molecules (parallel to the c axis of the crystal)
are stabilized by H-bonding interactions that involve the aqua ligand of Cu1 and the exocyclic
amino group of Hade as H-donors (Fig. 3, bottom). The non-coordinated water molecule seems
to play a key role providing an additional stabilization to these chains via H-bonding
interactions connecting the O2(aqua) atom of Cu2 and an O-carboxylate atom as H-acceptors.
Finally, the 3D architecture is accomplished by hydrophobic interactions between the benzyl
arms of BCBC ligands.
<Insert Figure 3>
3.2.2. Ternary complex with BCBC and bridging Hdap as coligand.
15

ACCEPTED MANUSCRIPT
Suitable crystals that lead to the structure of [Cu2(BCBC)2(µ2-N7,N9-H(N3)dap)(H2O)2]·4H2O
(2) were recently obtained (Fig. 4, top). The two metal centers are achiral and have the
stereochemical descriptor OC-6-32. They both exhibit an asymmetrically elongated octahedral
coordination, type 4+1+1, although with some differences. In the environment of the Cu1 ion
the Cu1-S1(thioether) bond is (2.658(3) Å), significantly shorter than the Cu1-O(apical aqua)
bond (2.917(5) Å). This Cu-O(aqua) bond can be considered a rather weak interaction since is
close to the sum of Van der Waals radii [(1.40 (Cu) + 1.50 (O) = 2.90 Å)]. However, the crystal
packing of 2 suggests a relevant impact of this massive but weak interaction in the final
structure, and therefore the Cu1 coordination could be described as a 4+1(S)+1(O)*
coordination (where the star indicates the weakness of the referred Cu···O(apical aqua)
interaction, sometimes so-called semi-coordination). If this interaction is ignored, then the Cu1
coordination could be simply seen as a 4+1 polyhedron. In contrast, the elongated octahedral
surrounding of Cu2 center is clearly defined as type 4+1+1, where the trans-apical distances are
Cu2-O(aqua) 2.704(6) Å < Cu2-S2 2.867(3) Å . The dinuclear molecular complex involves one
Hdap ligand displaying a µ2-N7,N9 bridging role. The diamino-purine ligand adopts the rare
tautomer H(N3)dap, with the tautomerizable proton on the less basic N heterocyclic atom!. That
allows the two Cu-N(Hdap) bonds [Cu-N9 (1.961(5) Å) and Cu-N7 (1.968(5) Å)] to be
reinforced by the corresponding intra-molecular interligand H-bonds [N3-H···O(coord.)
(2.732(7) Å, 139.9º) and N6-H···O(coord.) (2.774(7) Å, 169.1º), respectively].
<Insert Figure 4>
The structure of this molecule raises two interesting questions. First, why is the metal binding
pattern µ2-N7,N9-H(N3)dap preferred rather than others having the most stable H(N9)dap
tautomer? For example, the µ2-N3,N7-H(N9) mode (never reported- see below). Second, why
are the Cu-S distances such different between both metal centers? The involvement of the N1
16

ACCEPTED MANUSCRIPT
donor of Hdap in metal-binding seems unlikely due to the influence of both C6 and C2
exocyclic amino groups, which in addition can be protonated in the lactamic form. Thus, the µ2-
N7,N9-H(N3)dap bridging mode seems a good alternative to minimize the hindrance on the N3
donor caused by the steric effects imposed by the exocyclic amino group on C2. Besides, as
noted above, the reported metal binding mode offers the possibility to assist the M-N3 and M-
N7 coordination bonds with intra-molecular H-bonds, which further stabilize the molecule. In
contrast to the large number of structures reported about metal complexes with neutral Hade,
only limited structural information on neutral Hdap and its metal complexes is available [22].
Surprisingly, no monodentate metal binging pattern has been described for neutral Hdap. Three
structures are known on Zn(II) ternary complexes with 2-Hdap ligand [23]. The polymers
[Zn2(μ2-N7,N9-H(N3)dap)(tp)2]n and {[Zn2(μ2-N7,N9-H(N3) p)(μ2-OH)(tm)] ·H2O}n (tm =
trimesicate(3-) ion) exhibit the same metal binding pattern here reported. As discussed for
compound 2, these two Zn(II) compounds also have the rare tautomer H(N3)dap and both Zn-
N7 and Zn-N9 bonds are appropriately reinforced by intra-molecular N6-H··O(coord.) and N3-
H··O(non-coord.) interactions. Besides, in the polymer {[Zn4(µ4-tr)(µ3-HO)(µ2-N3,N9-
H(N7)dap)(µ2-tr)2]·2H2O}n (tr = terephthalate(2-) ion) a different tautomer is observed where
only the Zn-N3 bond cooperates with a N-H···O(coord.) interligand interaction.
Concerning the different Cu-S(thioether) bond distances, they might be related to crystal
packing forces. In the crystal of 2, centro-symmetric pairs of complex molecules are built by a
rather n π,π-stacking interactions between the six-membered rings of Hdap ligands [inter-
centroid distance dc-c 3.43 Å, inter-planar distance dπ,π 3.31 Å, inter-planar dihedral angle
w n k n α= 0º n pp n n w n k n β = γ = 15.4º] (SD3.2).
F ymm y n (α = 0º) h m k n n p . H n , the short
interplanar distance (also the stacking distance) and the low values of splitting angles (15º) are
related to the efficiency of the ‘ n -parallel k n ’ H p n . Note that their five-
membered rings are oriented in opposite sense. The stability of these pairs of molecules in 2 is
17

ACCEPTED MANUSCRIPT
additionally reinforced by two symmetry related H-bonds involving a N6-H bond of one
molecule and the O2-aqua ligand of other molecule as H-acceptor [3.009(8) Å, 137.2º]. In
2002, our research group reported the structure of the complex [Cu2(NBzIDA)2(µ2-N7,N9-
H(N3)ade)(H2O)2]·3H2O [24] (NBzIDA = N-benzyliminodiacetate(2-) ion) where adenine also
used the rare tautomer H(N3)ade and the aforementioned binding metal pattern. However, in
this case ny π,π-stacking interaction is observed in the crystal. The crystal of this NBzIDA-
Hade derivative was only built as a 3D H-bonded array of individual dinuclear complexes and
lattice water molecules while the crystal of compound 2 is constructed as a H-bonded
framework of ‘ k ’ pairs of dinuclear complexes and non-coordinated water molecules. In
fact, the crystal of 2 builds H-bonded double stranded chains of pairs of molecules, running
parallel to the a axis (Fig. 4, bottom). Further H-bonds (involving solvent water) connect these
chains in a 2D framework (parallel to ab plane). Then, only hydrophobic interactions between
the benzyl arms of BCBC ligands connect these layers (along the b axis), achieving the 3D
network.
3.3. Ternary copper(II) complexes with the binucleating and tripodal tetradentate chelator TCC.
The TCC ligand has proved to act as a binucleating ligand for copper(II) in the three compounds
with known structure [9-11]. Two of them have imidazole coligands [10,11] and one contains
h ‘ m ’ tautomer 5-methylimidazole (the most stable isomer of 4/5-imidazole) [11]. On this
basis, we have attempted to obtain related compounds with Hade or Hdap as N-heterocyclic
coligands. All our attempts to crystallize a ternary Cu(II)-TCC-Hdap compounds were
unsuccessful.
3.3.1. Ternary complex with TCC and Hade as coligands
According to the elemental analysis of a freshly prepared sample, compound 3 crystallizes with
the formula [Cu2(µ2-TCC)(Hade)2(H2O)2]·10H2O. However, the thermal stability of this
18

ACCEPTED MANUSCRIPT
decahydrate is rather low. Under air-dry flow, the decahydrate spontaneously loses water being
transitionally stabilized as [Cu2(µ2-TCC)(Hade)2(H2O)2]·4H2O (SD5.3). The overlapped loss of
the remaining solvent molecules and aqua ligands (25-150 ºC) is followed by the pyrolysis of
the organic ligands, which produces a significant variety of gases, including the three N-oxides
N2O, NO and NO2, and SO2. As referred in Section 2.3, the contribution of non-coordinated
water molecules could not be identified by X-ray diffraction. In compound 3, each dinuclear
complex molecule is centro-symmetric and achiral (Fig. 5, top). The stereochemical descriptor
of the copper(II) centers is OC-6-32. In the crystal, equimolar amounts of two symmetry-
related complex molecules coexist, differing in the Δ (79.55º) or Λ (-79.55º) values of the C-S-
S-C torsion angle in TCC. As expected [9-11], the octadentate-TCC ligand plays a 2-bridging
role, with each half-TCC acting as a tripodal-tetradentate chelator for each metallic center. The
six-coordination of the copper(II) center is satisfied by a tridentate iminodiacetate moiety (in
mer-NO2 conformation), one S-disulfide atom of TCC, the N7 donor of the stable H(N9)ade
tautomer and the O-aqua ligand. The S-disulfide and O-aqua donors lie in the trans-apical/distal
coordination sites. The Cu1-N7(Hade) bond (1.953(11) Å) is assisted by an intra-molecular N6-
H6B···O21(coord. carboxylate; 2.858(14) Å, 172.0º) interligand interaction. The Cu-S bond
(2.933(3) Å) corresponds to the longest coordination distance of the 4+1+1 Cu(II) surrounding.
The metal binding of Hade ligands does not influence either the chelating role of the TCC
ligand or the configuration of the IDA moiety as compared to the binary compound {[Cu2(µ2-
TCC)(H2O)2]·4H2O}n [9] or the related ternary complexes with imidazole [10] or 5-
methylimidazole [11].
In the crystal of 3, the dinuclear molecules build ribbons with inter-molecular π,π-stacking
interactions (SD3.3) between pairs of ‘anti-parallel’ Hade ligands [inter-planar distance dπ,π
~3.26 Å, inter-centroid distance dc-c 3.55 Å, inter-planar dihedral angle between stacked Hade
n α = 2.22º n pp n n w n k n β γ = 22.5º or 24.1º]. This
interaction represents the stacking of the five-membered ring of a Hade ligand with the six-
19

ACCEPTED MANUSCRIPT
membered ring of the adjacent one and vice versa (Fig. 5, bottom). Such anti-parallel π,π-
stacking interaction also builds pairs of complex molecules in the compound
[Cu(MIDA)(Hade)(H2O)]·H2O (MIDA = N-methyliminodiacetate(2-) ligand) [20]. The stability
of the ribbons in 3 is also supported by symmetry related pairs of inter-molecular H-bonding
interactions involving the apical aqua ligands and the N3#2 donors of a neighboring Hade
ligands [O1-H1B···N3 (2.82(2) Å, 162.5º, symmetry code #2 –x+1/2, -y+1/2, -z+2)]. Parallel
ribbons build a 3D H-bonded framework probably by the implication of solvent molecules.
< Insert Figure 5 >
3.3.2 Complexes with bidentate aromatic α,α’-diimines as coligands
It is known that the iminodiacetate chelate yields polymeric binary copper(II) compounds
[Cu(IDA)(H2O)2]n [25] or [Cu(IDA)(H2O)]n [26] where the chelator exhibits a fac-
NO+O(apical/distal) conformation. On the other hand, other IDA-like binary chelates with R-
2-
N(CH2COO)2 (R = non-coordinating alkyl [27-29]) show a mer-NO2 conformation. The
previously referred Cu(II) complexes with TCC in absence [9] or presence [10,11] of a
monodentate N-heterocyclic coligand (including now compound 3), feature the mer-NO2
iminodiacetate conformation. In the CSD database, there are many examples of Cu(II)
complexes with general formula [Cu(IDA)(L)(H2O)], where L is an unidentate N-heterocyclic
coligand, exhibiting the same conformation, i.e. including with imidazole [30], 2-aminopurine
[31], hypoxanthine [32] or acyclovir [33] as coligands. This mer-NO2 conformation might favor
the NO2S(apical)-tripodal tetradentate role of each half-TCC in the known compounds (and
perhaps also BCBC compounds 1 and 2). In contrast, the fac-NO+O(apical/distal) IDA
conformation is typically observed in complexes [Cu(IDA)(L-2)] where L-2 represents either
w m n n H m n [34] n m α,α’-diimine (phen or related ligands
[35-37]),
as
well
as
N(aliphatic),N(heterocyclic)-bidentate
20
ligands
(i.e.
2-

ACCEPTED MANUSCRIPT
aminomethylbenzymidazole [38]) or N(heterocyclic),O(amide)-bidentate ligands (as α-
pyridincarboxamide [39]).
All our attempts to bind two Hade or Hdap ligands to the same Cu(II) center in TCC or BCBC
chelates resulted unsuccessful. Therefore, we synthesized and studied novel ternary Cu(II)
mp x w h T n w n α,α’-diim n n : 2,2’-bipyridine (compound 4,
SD3.4 and SD3.5) and 1,10-phenanthroline (compound 5). Our goal is to discern if the IDA-
moieties of each half-TCC ligand are able to drive the fac-NO+O(apical/distal) conformation in
this kind of Cu(II) compounds and, in such case, if the S-disulfide atoms of TCC could still
remain linked to the metal centers. The molecular structures of the complexes 4 (Fig. 6) and 5
(Fig. 7) reveal that each half-TCC ligand acts efficiently as a tripodal-tetradentate chelator for
each copper(II) center. Interestingly, the TCC-IDA-like moiety adopts now a fac-
NO+O(apical/distal) conformation and this occurs maintaining the integrity of the Cu-
S(disulfide) bonds. Hence, the molecular structure of compounds 4 and 5 prove for the first time
the NO+O(apical)+S(apical)-tripodal tetradentate role of each half-TCC ligand.
Compound 4 crystallizes in the space group C2. The asymmetric unit has two non-equivalent
molecules (1 and 2) (Fig. 6). The Cu1 and Cu2 centers of molecule 1 exhibit an absolute
configuration of the type A (that is the four closets donors [O >N-bpy, N-bpy >N-TCC] in
counter-clockwise direction) and the dihedral angle C-S-S- (74.77º) h Δ n n. In
contrast, the Cu3 and Cu4 centers of molecule 2 exhibit an absolute configuration of the type C
(with the four closets donors [O >N-bpy, N-bpy >N-TCC] clockwise direction) and the dihedral
angle C-S-S-C (-75.35º) p Λ n n. No π,π-stacking interactions are observed.
The crystal packing is mainly built thanks to extended H-bonded networks, involving water
molecules as H-donors. The solvent occupies part of the voids (37.8% of crystal volume)
between the complex molecules.
The asymmetric unit of the crystal of compound 5 also has two non-equivalent molecules (Fig.
7, top) with Cu1 and Cu2 centers in molecule 1 and Cu3 and Cu4 centers in molecule 2. The
21

ACCEPTED MANUSCRIPT
metallic centers within each of these two molecules have the same absolute configuration (C for
molecule 1 and A for molecule 2). These two non-equivalent molecules also differ in the
dihedral angle C-S-S-C: -79.22 º (Λ) m 1 n +77.91º (Δ) m 2. In the
crystal, pairs of molecules are associated by a π,π-stacking interaction between the C6-phen
rings (SD3.5). The structural parameters are dπ,π=3.29 Å, dc-c = 3.37 Å, α = 0º n β = γ =
12.75º for stacked phen ligand linked to two Cu1 centers and dπ,π=3.27 Å, dc-c=3.55 Å, α =0º, β
= γ = 23.16º for phen ligand bonded to two Cu3 centers. This means that (a) only a half of the
phen ligands are involved in π,π-stacking interactions (see Fig. 7, bottom) and (b) the stacking is
strong (with lower v pp n n , β n γ, w n p phen ligands linked to
Cu1 centers). The crystal of the compound 5 consists of a 3D network built by contribution of
many H-bonds involving the non-coordinated water molecules, hosted in voids (35.6 % total
volume crystal) between the tetranuclear motifs.
< Insert Figure 6 and 7 >
3.4. The copper(II)-S(tioether) bond in Cu(II)-BCBC and related compounds.
In a recent paper by J. Reedijk [3] about the plasticity of the copper-thioether bonds, the
relevance of the Cu-S(thioether) bond in a variety of biological and bio-mimetic systems is
emphasized. It is clear that the borderline Cu(II) atom is a poorer binding center than the soft
Cu(I) atom for the soft S-thioether donor. In bio-mimetic systems, the Cu-S(thioeter) bond
distance ranges between 2.4-3.2 Å for Cu(I) or Cu(II), with a probabilistic peak around 2.8 Å.
The averaged CuII-S(thioether) distance for compounds 1 and 2 in this work is ~2.83 Å. It has
been argued that the shortest CuII-S(thioether) distances are found in cases where this bond is
oriented towards the magnetic orbital of the metal atom. Nevertheless, in compounds 1 and 2
the copper(II) is in a elongated octahedral or square base pyramidal coordination and the S-
thioether atom in an apical coordination site. In these compounds, with a dx2-y2 ground state, the
22

ACCEPTED MANUSCRIPT
CuII-S(thioether) is oriented to the filled dz2 orbital. The rather short CuII-S(thioether) bond in
compounds 1 and 2 would be related to the implication in the tetradentate chelating role of the
BCBC ligand. The presence of the S-benzyl arm does not represent a negative influence. The
absence of substituents on the C-atoms involved in chelate rings precludes steric constraints.
That seems to apply for the dinuclear chelate [[Cu2(µ2-TBENT)(H2O)2]·3H2O}2 [40] (TBENT =
S(C2H4N(CH2COO)2)2. This polymer has asymmetric dinuclear motifs with a bridging S-
thioether atom which gives Cu-S bond distances of 2.751(1) and 2.829(1) Å.
3.5. The copper(II)-S(disulfide) bond in Cu(II)-TCC related chelates.
Table 2 summarizes data regarding the coordination environment of Cu(II) as well as the Cu-S
and S-S bond distances in complexes derived from TCC and characterized by X-ray
crystallography. To the best of our knowledge, the chelating ligand TCC always acts as bridging
ligand between two metal centers. The Cu(II)-TCC binary compound is a polymer while all
related ternary Cu(II) complexes are molecular. In these complexes, each half-TCC ligand acts
as a NO2S-tripodal tetradentate chelator, displaying the IDA-mer-NO2 conformation and the S-
disulfide atom in an apical/distal coordination site of the copper(II) center. In contrast, in those
ternary compounds with α,α’-aromatic diimines as coligand, the IDA moiety is moved to a fac-
NO+O(apical/distal) conformation. In this case, the two metal-glycinate chelate rings fall nearly
perpendicular and one of the O-carboxylate donor is coordinated in the trans-apical position to
the S(thioether) donor. This is a consequence of the preference of bpy and phen to occupy two
among the four closest sites of the Cu(II) surrounding. Noticeably, the induction of such fac-
NO+O(apical/distal) conformation does not modify the tripodal tetradentate role of the half-
TCC ligand. All ternary Cu(II) compounds have Cu-S(disulfide) bond distances (2.798-2.981
Å) longer than that observed for the corresponding binary compound (2.721 Å) (see Table 2).
The Cu-S bond distance in the penta-coordinated ternary Cu(II) complex having H5Meim
(2.798 Å) is shorter than the corresponding Cu-S average bond distance (2.911 Å) of ternary
23

ACCEPTED MANUSCRIPT
complexes with TCC. This fact should be essentially attributed to the penta-coordination of the
metal center, opposite to the distorted octahedral Cu(II) coordination shown in the rest of the
ternary complexes. The S-S distance stays in the range of 2.036 – 2.051 Å. The interatomic
Cu···Cu distance, between Cu(II) atoms chelated by the same TCC ligand, is significantly
increased from the binary chelate (4.36 Å) to the ternary compounds (5.32 – 5.97 Å).
From a biological point of view, the Cu-S(disulfide) bonds in TCC complexes are shorter than
those reported for [Cu2(D-pds)2(H2O)2]·7H2O [7], where each disulfide group is monodentate
(Cu-S 3.057(7) or 3.138(7) Å), and Na4[Cu2(GSSG)]·6H2O [8], where the disulfide group acts
as bidentate, giving two unequal Cu-S bonds (3.16(1) or 3.28(1) Å). In 2008, T. Konno et al.
[41] re-determined the structure of [Cu2(D-pds)2(H2O)2]·7.5H2O and also reported the
sophisticated structures of two hydroxo-bridged hepta-copper(II) clusters with double-cubane
frameworks, sharing a corner and supported by three D-pds ligands: [Cu7(µ3-OH)6(µ3-X)2(D-
pds)3] (X or Cl or Br). Th x ‘ -m m ’ h -cubane framework in the
chloride derivative give Cu-S(disulfide) bonds with an averaged distance of 3.314(2) Å in a
distorted six-coordinated surrounding. Additionally, T. Konno et al. [42] have studied the
influence of various N,N-coligands (such as bpy or phen) in the so-called auto-oxidation (on air)
of thiol-containing amino acids (L-cysteine or D-penicillamine) to their corresponding disulfide
forms [L-cystine (L-cysti) or D-penicillamine disulfide (D-pds)] in copper(II) ternary systems.
X-ray diffraction results led to the structure of two novel mixed-ligand metal complexes
according to the formulas [Cu4(bpy)4(µ4-L-cysti)](ClO4)4·1.5H2O and [Cu2(bpy)2(H2O)2(µ2-D-
pds)](PF6)2·2MeOH. However, in contrast to that here reported for compounds 4 and 5, the S-
disulfide atoms in these complexes are not linked to the Cu(II) atoms, which exhibit square-
based pyramidal coordination type 4+1 (with two N-bpy and one O plus the N donor of a
glycinate moiety as basal donors and the O-aqua or one O-carboxylate as apical ligand).
4. Concluding remarks
24

ACCEPTED MANUSCRIPT
In past decades, a relevant number of research papers, reviews and patents have been reported
regarding the applications of S-chelating agents, covering three main aspects: (a) The use of S-
containing chelators for the treatment of heavy-metal intoxications; (b) The role of sulfur-amino
acids and related chelators in bio-metal-disturbances, such as Wilson disease; (c) The use of
N,O,S-chelators as carriers of 99m-Tc and other radioactive isotopes [2]. Most of these results
were developed under the guidance of semi-empirical structure-function relationships. Our
structural results aim to understand the rationale for the molecular recognition of these
molecules, for instance the tripodal-tetradentate role of this kind of ligands. It should be kept in
mind that biological systems generally represent reductive media, where disulfide groups can be
reductively broken and then oxidized, reconstituting the disulfide bond. Indeed, the blue-
greenish compound {[Cu2(µ2-TCC)(H2O)2]·4H2O}n [9] was first synthesized by reaction
between metallic copper and H4TCC acid, through a violet Cu(I)-sulfide intermediate, which
exposed on air undergoes an oxidation reconstructing the disulfide group and TCC ligand.
In this work, we have proved that the dinuclear chelate Cu2(TCC) is able to bind adenine. In this
compound, each half-TCC features the NO2+S(apical disulfide) tetradentate role and the Cu-
N7(Hade) bond is further assisted by an intra-molecular N6-H···O(coordinated) interligand
interaction. The NO2+S(apical thioether) chelating role has also been shown for the dinuclear
chelate Cu2(BCBC)2 with 2-N3,N7-H(N9)ade and 2-N7,N9-H(N3)dap N-coligands. In these
complexes all Cu-N(purine) bonds act in cooperation with appropriate intra-molecular
(purine)N-H···O(coordinated carboxylate) interligand interactions. The tripodal-tetradentate
roles of a half-TCC and of BCBC ligands are believed to be related to (a) their N-branched
skeletal topology (in particular, without C-side chains in their acetate arms, as in the amino
acids cysteine and penicillamine or their corresponding disulfides) and (b) the ability of these
polydentate ligands to build five-membered N,S-chelate rings. The stability of these rings is also
proved by the structures of [Cu2(TCC)(bpy or phen)2] complexes where each half-TCC ligand
exhibits tetra-denticity in the NO+O(apical)+S(apical) conformation. This is in contrast to the
25

ACCEPTED MANUSCRIPT
behavior previously observed on related ternary complexes derived from L-cystinate or D-
penicilaminate disulfide ligands, which does not build Cu(II)-S disulfide bonds.
Acknowledgements
Financial support from Research Group FQM-283 (Junta de Andalucía) and MICINN Spain
(ProjectMAT2010-15594) kn w . Th p j ‘F í z ón,
CONSOLIDER INGENIO-2010’ p v X-ray structural facilities for this work. Financial
support from ERDF Funds and Junta de Andalucía to acquire the FT-IR spectrophotometer
Jasco 6300 is acknowledged. ADM gratefully acknowledges a postdoctoral fellowship from
Fundación Ramón Areces. HEB gratefully acknowledges a postdoctoral contract from the
University of Granada (Plan Propio de Investigación 2013).
Appendix A. Supplementary data
Structural information on H4TCC·3H2O (SD1), NMR and FT-IR of the S-containing chelating
ligands in acid forms (SD2) is provided. Additional structural details on the copper(II)
compounds 1-5 is available in SD3. As an example, only the spectral and thermogravimetric
graphs of compounds 1 (SD4) and 3 (SD5) are available. For further information about the
spectral (FT-IR and UV/visible) or thermo-gravimetric properties carried out to all the reported
compounds, please contact with the corresponding author. Supplementary data to this article can
be found n n :….
26

ACCEPTED MANUSCRIPT
References
[1] W. Kaim, B. Schwederski, A. Klein, Bioinorganic chemistry. Inorganic elements in the
chemistry of life: an introduction and guide, 2nd edition, Wiley, 2013.
[2] E. Chiotellis, C.I. Stassinopoulou, A. Varvarigou, H. Vavouraki, J. Med. Chem. 25 (1982)
1370-1374.
[3] J. Reedijk, J. Inorg. Biochem. 115 (2012) 182-185.
[4] T. Ohta, T. Tachiyama, K.Yoshizawa, T. Yamabe, Tetrahedron Letters 41 (2000) 2581-
2585.
[5] P.J.M.W.L. Birker, H.C. Freeman, J. Chem. Soc., Chem. Comm. 9 (1976) 312-313.
[6] H.J. Schugar, C.-C. Ou, J.A. Thich, J.A. Potenza, T.R. Felthouse, M.S. Haddad, D.N.
Hendrickson, W. Furey, R.A. Lalancette, Inorg. Chem. 19 (1980) 543-552.
[7] J.A. Thich, D. Mastropaolo, J. Potenza, H.J. Schugar, J. Am. Chem. Soc. 96 (1974) 726-731.
[8] K. Miyoshi, Y. Sugiura, K. Ishizu, Y. Iitaka, H. Nakamura, J. Am. Chem. Soc. 102
(1980) 6130-6136.
[9] N.-H. Dung, B. Viossat, J.M. González-Pérez, J. Niclós-Gutiérrez, Acta Crystallogr. C41
(1985) 1739-1742.
[10] J.M. González-Pérez, N.-H. Dung, J. Niclós-Gutiérrez, B. Viossat, A. Busnot, M.L.
Vicente-Gelabert, Inorg. Chim. Acta 166 (1989) 115-122.
[11] J. Niclós-Gutiérrez, E. Abarca-García, B. Viossat, N.-H. Dung, A. Busnot, J.F. Hemidy,
Acta Crystallogr. C49 (1993) 19-22.
[12] C.G. Agoston, K. Varnagy, A. Benyei, D. Sanna, G. Micera, I. Sovago, Polyhedron 19
(2000) 1849-1857.
[13] A.K. Patra, M. Nethaji, A.R. Chakravarty, J. Inorg. Biochem. 101 (2007) 233-244.
[14] Bruker, SMART, SAINT. Area detector control integration software. Bruker Analytical X-
ray Instruments, Inc., Madison, Wisconsin (1999).
27

ACCEPTED MANUSCRIPT
[15] BRUKER APEX2 Software. Bruker AXS Inc., Madison, Wisconsin, USA (2010)
V2010.11.
[16] G.M. Sheldrick, SADABS. Program for empirical absorption correction of area detector
data. University of Göttingen, Germany (2008).
[17] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112–122.
[18] A.L. Spek, PLATON. A multipurpose crystallographic tool. Utrecht University, Utrecht,
The Netherlands (2010).
[19] C.F. Macrae, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock, L. Rodríguez-
Monge, R. Taylor, J. Van de Streek, P.A. Wood, J. Appl. Crystallogr. 41 (2008) 466-470.
[20] E. Bugella-Altamirano, D. Choquesillo-Lazarte, J.M. González-Pérez, M.J. Sánchez-
Moreno, R. Marín-Sánchez, J.D. Martín-Ramos, B. Covelo, R. Carballo, A. Castiñeiras, J.
Niclós-Gutiérrez, Inorg. Chim. Acta 339 (2002) 160–170.
[21] M.P. Brandi-Blanco, D. Choquesillo-Lazarte, A. Matilla-Hernández, J.M. González-Pérez,
A. Castiñeiras, J. Niclós-Gutiérrez, J. Inorg. Biochem. 127 (2013) 211-219.
[22] A. Domínguez-Martín, M.P. Brandi-Blanco, A. Matilla-Hernández, H. El Bakkali, V.M.
Nurchi, J.M. González-Pérez, J. Niclós-Gutiérrez, Coord. Chem. Rev. 257 (2013) 2814–2838.
[23] E.-C. Yang, Y.-N. Chang, H. Liu, Z.-C. Wang, X.-J. Zhao, Cryst. Growth Des. 9 (2009)
4933-4944.
[24] P.X. Rojas-González, A. Castiñeiras, J.M. González-Pérez, D. Choquesillo-Lazarte, J.
Niclós-Gutiérrez, Inorg. Chem. 41 (2002) 6190–6192.
[25] M.J. Román-Alpiste, J.D. Martín-Roamos, A. Castiñeiras-Campos, E. Bugella-Altamirano,
A.G. Sicilia-Zafra, J.M. González-Pérez, J. Niclós-Gutiérrez, Polyhedron 18 (1999) 3341-3351.
[26] Q. Zhong, Y.-H. Wang, X.-T. Zhang, Acta Crystallogr. E67 (2011) m1528.
[27] T.N. Polynova, N.V. Dvortsova, T.N. Filippova, M.A. Porai-Koshits, Y.V. Nekrasov,
Koord. Khim. 12 (1986) 1484-1489.
28

ACCEPTED MANUSCRIPT
[28] P.X. Rojas-González, D. Choquesillo-Lazarte, J.M. González-Pérez, S.A. Ruiz-García, R.
Carballo, A. Castiñeiras, J. Niclós-Gutiérrez, Polyhedron 22 (2003) 1027- 1037.
[29] R. Fernández-Piñar, C. Sánchez de Medina-Revilla, A. Domínguez-Martín, M.P. Brandi-
Blanco, D. Choquesillo-Lazarte, J.M. González-Pérez, J. Niclós-Gutiérrez, Z. Anorg. Allg.
Chem. 633 (2007) 2658-2666.
[30] A. Castiñeiras, J.M. Tercero, A. Matilla, J.M. González, A.G. Sicilia, J. Niclós, J. Coord.
Chem. 35 (1995) 61- 72.
[31] A. Domínguez-Martín, D. Choquesillo-Lazarte, J.M. González-Pérez, A. Castíñeiras, J.
Niclós-Gutiérrez, J. Inorg. Biochem. 105 (2011) 1073-1080.
[32] D.K. Patel, D. Choquesillo-Lazarte, A. Domínguez-Martín, M.P. Brandí-Blanco, J.M.
González-Pérez, A. Castíñeiras, J. Niclós-Gutiérrez, Inorg. Chem. 50 (2011) 10549-10551.
[33] M.P. Brandi-Blanco, D. Choquesillo-Lazarte, A. Domínguez-Martín, J.M. González-
Pérez, A. Castiñeiras, J. Niclós-Gutiérrez, J. Inorg. Biochem. 105 (2011) 616-623.
[34] N.-H. Dung, B. Viossat, A. Busnot, A.G. Sicilia-Zafra, J.M. González-Pérez, J. Niclós-
Gutiérrez, Inorg. Chim. Acta 169 (1990) 9-12.
[35] Z.A. Siddiqi, M. Shahid, M. Khalid, S Kumar, Eur. J. Med. Chem. 44 (2009) 2517-2522.
[36] S.K. Seth, B. Dey, S. Mukhopadhyay, J. Mol. Struct. 973 (2010) 81-88.
[37] B. Selvakumar, V. Rajendiran, P.U. Maheswari, H. Stoeckli-Evans, M. Palaniandavar, J.
Inorg. Biochem. 100 (2006) 316-330.
[38] I.S. de la Cueva, A.G. Sicilia, J.M. González, E. Bugella, A. Castiñeiras, J. Niclós-
Gutiérrez, React. Funct. Polym. 36 (1998) 211-216.
[39] E. Bugella-Altamirano, J.M. González-Pérez, D. Choquesillo-Lazarte, J. Niclós-Gutiérrez,
A. Castiñeiras-Campos, Z. Anorg. Allg. Chem. 626 (2000) 930-936.
[40] J.B. Berg, K.O. Hodgson, Inorg. Chem. 25 (1984) 1800-1803.
[41] A. Igashira-Kamiyama, J. Fujioka, S Mitsunaga, M. Nakano, T. Kawamato, T. Konno,
Chem. Eur. J. 14 (2008) 9512-9515.
29