Palladium(II) oxalato complexes involving N6-(benzyl)-9-isopropyladenine-based N-donor carrier ligands: Synthesis, general properties, 1H, 13C and 15N{1H} NMR characterization and in vitro cytotoxicity

Article abstract of DOI:10.1016/j.ica.2010.01.035

Reactions of potassium bis(oxalato)palladate dihydrate, K₂[Pd(ox)₂]·2H₂O, with two molar equivalents of N6-(benzyl)-9-isopropyladenine-based organic molecules (L_1-7), i.e. 2-chloro-N6-(2-methoxybenzyl)-9-isopropyladenine (L₁), 2-chloro-N6-(3-methoxybenzyl)-9-isopropyladenine (L₂), 2-chloro-N6-(3,5-dimethoxybenzyl)-9-isopropyladenine (L₃), 2-(1-ethyl-2-hydroxyethylamino)-N6-(benzyl)-9-isopropyladenine (L₄), 2-(1-ethyl-2-hydroxyethylamino)-N6-(2-methoxybenzyl)-9-isopropyladenine (L₅), 2-(1-ethyl-2-hydroxyethylamino)-N6-(3-methoxybenzyl)-9-isopropyladenine (L₆) and 2-(1-ethyl-2-hydroxyethylamino)-N6-(4-methoxybenzyl)-9-isopropyladenine (L₇), provided a series of seven palladium(II) oxalato (ox) complexes of the general formula [Pd(ox)(L_1-7)₂]·nH₂O (1-7; n = 0 for 4, 5 and 7, 3/4 for 1 and 2, 1 for 6, and 3 for 3). The compounds were characterized by elemental analysis, IR, Raman, ¹H, ¹³C and ¹⁵N{¹H} NMR spectroscopy, ESI+ mass spectrometry, molar conductivity and TG/DTA thermal analysis. The geometry of [Pd(ox)(L₂)₂] (2) was optimized on the B3LYP/6-311G*/LANL2DZ level of theory. The complexes 4-7 represent the first palladium(II) oxalato complexes with a PdN₂O₂ donor set, which involve highly potent purine-based cyclin-dependent kinase (CDK) inhibitors (L_4-7) as carrier N-donor ligands. The selected complexes 1, 3-5 and 7 were tested by an MTT assay for their in vitro cytotoxic activity against human osteosarcoma (HOS) cancer cell line. The highest activity was found for the complexes 5 (IC₅₀ = 34.9 μM) and 7 (IC₅₀ = 39.2 μM).

Full text of DOI:10.1016/j.ica.2010.01.035

Inorganica Chimica Acta 363 (2010) 1469–1478
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Palladium(II) oxalato complexes involving N6-(benzyl)-9-isopropyladenine-based
1
N-donor carrier ligands: Synthesis, general properties, H, 13C and 15N{1H} NMR
characterization and in vitro cytotoxicity
*
ˇ
ˇ
Pavel Štarha, Igor Popa, Zdenek Trávnícek
ˇ
Department of Inorganic Chemistry, Faculty of Science, Palacky´ University, Tr. 17. listopadu 12, CZ-771 46 Olomouc, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Reactions of potassium bis(oxalato)palladate dihydrate, K2[Pd(ox)2]ꢀ2H2O, with two molar equivalents of
N6-(benzyl)-9-isopropyladenine-based organic molecules (L1–7), i.e. 2-chloro-N6-(2-methoxybenzyl)-9-
isopropyladenine (L1), 2-chloro-N6-(3-methoxybenzyl)-9-isopropyladenine (L2), 2-chloro-N6-(3,5-dime-
thoxybenzyl)-9-isopropyladenine (L3), 2-(1-ethyl-2-hydroxyethylamino)-N6-(benzyl)-9-isopropyladenine
(L4), 2-(1-ethyl-2-hydroxyethylamino)-N6-(2-methoxybenzyl)-9-isopropyladenine (L5), 2-(1-ethyl-2-
hydroxyethylamino)-N6-(3-methoxybenzyl)-9-isopropyladenine (L6) and 2-(1-ethyl-2-hydroxyethyl-
Received 7 December 2009
Received in revised form 15 January 2010
Accepted 22 January 2010
Available online 29 January 2010
Keywords:
Palladium(II) complexes
Oxalate
Adenine derivatives
CDK inhibitors
In vitro cytotoxicity
amino)-N6-(4-methoxybenzyl)-9-isopropyladenine (L7), provided a series of seven palladium(II) oxalato
3
(ox) complexes of the general formula [Pd(ox)(L1–7)2]ꢀnH2O (1–7; n = 0 for 4, 5 and 7,
for 1 and 2, 1
4
for 6, and 3 for 3). The compounds were characterized by elemental analysis, IR, Raman, 1H, 13C and
15N{1H} NMR spectroscopy, ESI+ mass spectrometry, molar conductivity and TG/DTA thermal analysis.
*
The geometry of [Pd(ox)(L2)2] (2) was optimized on the B3LYP/6-311G /LANL2DZ level of theory. The com-
plexes 4–7 represent the first palladium(II) oxalato complexes with a PdN2O2 donor set, which involve
highly potent purine-based cyclin-dependent kinase (CDK) inhibitors (L4–7) as carrier N-donor ligands.
The selected complexes 1, 3–5 and 7 were tested by an MTT assay for their in vitro cytotoxic activity
against human osteosarcoma (HOS) cancer cell line. The highest activity was found for the complexes 5
(IC50 = 34.9 lM) and 7 (IC50 = 39.2 lM).
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
single-crystal X-ray analysis. The mentioned complexes I and II
have the tetra-coordinated central Pd(II) ion which is surrounded
by one bidentate-coordinated oxalate dianion and by two mono-
dentate bonded adenine-based molecules (LI or LII) in a PdN2O2 do-
nor set. Moreover, these complexes were tested by a calcein
acetoxymethyl (AM) assay for their in vitro cytotoxic activity
against breast adenocarcinoma (MCF-7) and chronic myelogenous
leukaemia (K562) human cancer cell lines. Two of the tested sub-
stances showed promising in vitro cytotoxicity (IC50 values of 6.2
The plant hormone N6-(benzyl)adenine [6-(benzylamino)-
purine] [1] and its derivatives were found to be suitable N-donor
ligands of transition metal complexes. In the field of palladium(II)
complexes, the cis-[PdCl2(L)2], trans-[PdCl2(L)2], [PdCl3(L+)], [PdCl2-
(H2O)(L)], [PdCl(H2O)2(Lꢁ)] and [Pd(ox)(L)2] types of compounds
were prepared in our laboratory (see lit. [2–5] and the reference
cited therein), where L, L+ and Lꢁ stand for an electroneutral, pro-
tonated, and deprotonated N6-(benzyl)adenine derivative, respec-
tively, and ox symbolizes an oxalate dianion.
Talking about the [Pd(ox)(L)2] compounds in more detail, five
complexes with 2-chloro-N6-(benzyl)-9-isopropyladenine, or its
analogues with the substituted benzyl group, have recently been
published [3]. The molecular and crystal structures of two
complexes involving 2-chloro-N6-(4-methoxybenzyl)-9-isopro-
pyladenine (LI; complex I) and 2-chloro-N6-(4-methylbenzyl)-
and 6.8
platinum-based anticancer drugs Cisplatin (IC50 = 10.9
Oxaliplatin (IC50 = 18.2 M). To our best knowledge, only the
[Pd(ox)(Hoen)2]ꢀ0.5H2O and [Pd(ox)(Clen)2] complexes (Hoen =
N,N0-bis(hydroxyethyl)ethylenediamine, Clen = N,N0-bis(chloro-
l
M), which is higher than those of the commercially used
l
M) and
l
ethyl)ethylenediamine) [6], besides the above-mentioned
[Pd(ox)(L)2] compounds prepared in our laboratory, were tested
for their in vitro cytotoxicity within a group of monomeric palla-
dium(II) oxalato complexes, however, these substances were inac-
tive against mice leukaemia (P388) cells. On the other hand, the
results obtained in the case of [Pd(ox)(L)2] showed that this type
of complexes represents a promising group of compounds in
9-isopropyladenine (LII; complex II), were determined by
a
* Corresponding author. Tel.: +420 585 634 352; fax: +420 585 634 954.
E-mail address: zdenek.travnicek@upol.cz (Z. Trávnícek).
ˇ
0020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.01.035

1470
P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
Scheme 1. The derivatives of N6-(benzyl)-9-isoropyladenine (L1–7) used for the preparation of the [Pd(ox)(L1–7)]ꢀnH2O palladium(II) oxalato complexes.
connection with their in vitro cytotoxicity. For comparison, other
types of biologically active palladium complexes have been re-
viewed in the literature [2,4].
2. Experimental
2.1. Starting materials
In this paper, we present results following from our ongoing re-
search of palladium(II) oxalate complexes involving N6-(benzyl)-
9-isopropyladenine-based N-donor carrier ligands. We prepared
and characterized seven [Pd(ox)(L)2]ꢀnH2O complexes of which
the compounds 1–3 represent analogues of recently reported pal-
ladium(II) oxalato complexes [3] varying in the substitution on a
benzene ring of the 2-chloro-N6-(benzyl)-9-isopropyladenine moi-
ety (L1–3; see Scheme 1). On the other hand, the complexes 4–7 in-
volve differently substituted type of N6-(benzyl)adenine
derivatives with 2-amino-1-butanol at the C2 position of a purine
ring instead of the chlorine atom, namely 2-(1-ethyl-2-hydroxy-
ethylamino)-N6-(benzyl)-9-isopropyladenine (Roscovitine, L4) [7]
and its benzyl-substituted analogues (L5–7; Scheme 1). It is known
that Roscovitine and its derivatives belong to the group of highly
potent cyclin-dependent kinase (CDK) inhibitors, and thus the pre-
sented complexes 4–7 represent the first palladium(II) oxalato
complexes with purine-based CDK inhibitors acting as N-donor
carrier ligands.
Based on the above-mentioned statements, we decided to carry
out an in vitro cytotoxicity screening of the prepared complexes
against human osteosarcoma cancer cell line (HOS). The obtained
results showed that the complexes 5 and 7, involving the potent
CDK inhibitors, have in vitro cytotoxicity comparable with Cisplatin,
as discussed below. These findings motivated us to evaluate deeply
the in vitro cytotoxicity of these complexes, and thus, the named
palladium(II) oxalato complexes are currently tested against a vari-
ety of human cancer cell lines, e.g. MCF-7, lung carcinoma (A549),
cervix epithelioid carcinoma (HeLa), ovarian carcinoma (A2780),
Cisplatin-resistant ovarian carcinoma (A2780cis) or malignant mel-
anoma (G-361).
Chemicals and solvents were purchased from the commercial
sources (Sigma-Aldrich Co., Acros Organics Co., Lachema Co. or Flu-
ka Co.) and they were used as received. Dimethyl sulfoxide (DMSO)
was dried using MgSO4.
Potassium bis(oxalato)palladate(II) dihydrate, K2[Pd(ox)2]ꢀ
2H2O, was prepared from potassium tetrachloropalladate(II),
K2[PdCl4], as formerly described [3,8]. Syntheses of 2-chloro-N6-
(2-methoxybenzyl)-9-isopropyladenine (L1), 2-chloro-N6-(3-meth-
oxybenzyl)-9-isopropyladenine (L2), 2-chloro-N6-(3,5-dimethoxy-
benzyl)-9-isopropyladenine (L3), 2-(1-ethyl-2-hydroxyethylamino)-
N6-(benzyl)-9-isopropyladenine (L4), 2-(1-ethyl-2-hydroxyethyl-
amino)-N6-(2-methoxybenzyl)-9-isopropyladenine (L5), 2-(1-ethyl-
2-hydroxyethylamino)-N6-(3-methoxybenzyl)-9-isopropyladenine
(L6) and 2-(1-ethyl-2-hydroxyethylamino)-N6-(4-methoxyben-
zyl)-9-isopropyladenine (L7) were inspired by several literature
sources, and their structures are depicted in Scheme 1 [9–12]. A
scheme of the synthetic pathway of the L1–7 compounds, as well
as the results of IR, Raman and NMR spectroscopies, are given in
Appendix A in Supplementary material.
3
3
2.2. Preparation of [Pd(ox)(L1)2]ꢀ 4H2O (1), [Pd(ox)(L2)2]ꢀ 4H2O (2),
[Pd(ox)(L3)2]ꢀ3H2O (3), [Pd(ox)(L4)2] (4), [Pd(ox)(L5)2] (5),
[Pd(ox)(L6)2]ꢀH2O (6) and [Pd(ox)(L7)2] (7)
The palladium(II) oxalato complexes 1–7 were prepared accord-
ing to a general synthetic procedure recently published and de-
picted in Scheme 2 [3]. Briefly, a K2[Pd(ox)2]ꢀ2H2O distilled water
solution (15 mL, 40 °C) was mixed together with an acetone solu-

P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
1471
Scheme 2. A schematic representation of the preparation pathway of the palladium(II) oxalato complexes 1–7.
tion (15 mL, 25 °C) of the appropriate organic molecule L1–7 in a 1:2
molar ratio. The product, which formed during two days of stirring
at the temperature of 40 °C, was filtered off, washed with hot
(5 mL) and cold (5 mL) distilled water and acetone (5 mL) and
dried in the air at 40 °C.
MS (methanol, m/z): [Pd(ox)(L3)3 + H]+ 1279.4, [Pd(ox)(L3)2 + H]+
918.6, [Pd(ox)(L3) + K]+ 596.8, [Pd(ox)(L3) + H]+ 557.9, [L3 + H]+
362.1. IR (Nujol; cmꢁ1): 560vs
cmꢁ1): 3137w, 3111w, 3057w
m
m
(Pd–O); 521s
(C–H)ar, 2982w, 2941w, 2838w
(C@N); 1539w, 1470s
(C–O)ar; 1156s (C–Cl); 560w
(Pd–N). Raman (cmꢁ1): 3333w
(N–H); 3137w,
(C–H)ar; 2985s, 2945s, 2838w (C–H)al; 1697s,
(C@O)ox; 1535w, 1486w (C@C)ar; 1166w (C–Cl); 560s
(Pd–O); 521w
(Pd–N). 1H NMR (DMF-d7, ppm): d 9.18 (t, 6.2,
m(Pd–N). IR (KBr;
m
m
m
(C–H)al; 1712vs, 1676s
m
(C@O)ox; 1619vs
m
3
1: Yield: 580 mg (67%). Anal. Calc. for C34H36N10O6Cl2Pdꢀ 4H2O:
(C@C)ar; 1378s
m
(C–O)ox; 1229w
m
m
C, 46.9; H, 4.3; N, 16.1. Found: C, 46.8; H, 4.5; N, 15.9%. m.p. 169–
171 °C (decomp.). KM (DMF solution, S cm2 molꢁ1): 5.8. ESI+ MS
(methanol, m/z) [Pd(ox)(L1)3 + H]+ 1189.5, [Pd(ox)(L1)2 + K]+
897.0, [Pd(ox)(L1)2 + Na]+ 880.9, [Pd(ox)(L1)2 + H]+ 858.8,
[Pd(ox)(L1) + K]+ 566.0, [Pd(ox)(L1) + Na]+ 549.8, [Pd(ox)(L1) + H]+
(Pd–O); 521w
m
m
m
3076w, 3014w
m
1657w
m
m
m
m
m
N6H, 1H), 8.76 (s, C8H, 1H), 6.72 (d, 2.4, C11H, C15H, 2H), 6.41 (t,
2.4, C13H, 1H), 4.82 (d, 6.2, C9H, 2H), 4.79 (sp, 6.8, C16H, 1H), 3.81
(s, C23H, C24H, 6H), 1.53 (d, 6.8, C17H, C18H, 6H). 13C NMR (DMF-
d7, ppm): d 165.59 (C25, C26), 161.67 (C12, C14), 155.14 (C6),
154.10 (C2), 150.36 (C4), 143.80 (C8), 141.65 (C10), 117.32 (C5),
106.48 (C11, C15), 99.75 (C13), 55.66 (C23, C24), 49.64 (C16), 45.39
(C9), 21.99 (C17, C18). 15N NMR (DMF-d7, ppm): d 231.2 (N1),
223.8 (N3), 185.8 (N9), 147.5 (N7), 100.4 (N6).
4: Yield: 320 mg (71%). Anal. Calc. for C40H52N12O6Pd: C, 53.2; H,
5.8; N, 18.6. Found: C, 52.7, H, 6.1; N, 18.5%. m.p. 164–166 °C (de-
comp.). KM (DMF solution, S cm2 molꢁ1): 0.3. ESI+ MS (methanol,
m/z): [Pd(ox)(L4)3 + H]+ 1257.0, [Pd(ox)(L4)2 + Na]+ 925.2,
[Pd(L4)2 + H]+ 813.1, [Pd(L4) + Na]+ 571.1, [Pd(L4) + H]+ 461.2,
527.9, [L1 + H]+ 332.1. IR (Nujol; cmꢁ1): 559vs
(Pd–N). IR (KBr; cmꢁ1): 3130w, 3113w, 3066w
2939w, 2836w (C–H)al; 1707vs, 1671s (C@O)ox
(C@N); 1539w, 1491s (C@C)ar; 1364vs (C–O)ox; 1242vs
(C–Cl); 560w
(Pd–O). 1H NMR (DMF-d7, ppm): d
m
(Pd–O), 518
(C–H)ar, 2981w,
1622vs
(C–
m
m
m
m
;
m
m
m
m
O)ar; 1163w
m
m
8.99 (t, 6.2, N6H, 1H), 8.76 (s, C8H, 1H), 7.40 (dd, 7.3, 1.7, C15H,
1H), 7.28 (tt, 7.9, 1.7, C14H, 1H), 7.05 (dd, 8.2, 1.1, C12H, 1H), 6.86
(tt, 7.5, 1.1, C13H, 1H), 4.88 (d, 6.2, C9H, 2H), 4.81 (sp, 6.8, C16H,
1H), 3.94 (s, C23H, 3H), 1.53 (d, 6.8, C17H, C18H, 6H). 13C NMR
(DMF-d7, ppm): d 165.57 (C25, C26), 157.95 (C11), 155.27 (C6),
154.25 (C2), 150.31 (C4), 143.85 (C8), 129.02 (C13), 128.72 (C15),
126.66 (C10), 120.98 (C14), 117.31 (C5), 111.14 (C12), 55.95 (C23),
49.67 (C16), 40.51 (C9), 21.97 (C17, C18). 15N NMR (DMF-d7, ppm):
d 230.1 (N1), 223.0 (N3), 185.8 (N9), 147.3 (N7), 97.5 (N6).
[L4 + H]+ 355.3. IR (Nujol; cmꢁ1): 559vs
m
m
(Pd–O); 524vs
(C–H)ar, 2965s, 2932s,
(C@O)ox 1610vs (C@N);
(C–O)ox; 1058s (C–O)al; 560w
(Pd–N). Raman (cmꢁ1): 3148w, 3057vs
(C–H)ar;
2979s, 2935vs, 2877s (C–H)al; 1692w, (C@O)ox; 1606vs (C–N);
1491s (C@C)ar; 561w
(Pd–O). 1H NMR (DMF-d7, ppm): d 8.43
m(Pd–N).
IR (KBr; cmꢁ1): 3131w, 3062w, 3030w
3
2: Yield: 590 mg (69%). Anal. Calc. for C34H36N10O6Cl2Pdꢀ 4H2O:
2875s
1545vs, 1493vs
(Pd–O); 523w
m
(C–H)al; 1706vs, 1674s
m
;
m
C, 46.9; H, 4.3; N, 16.1. Found: C, 46.9; H, 4.4; N, 16.5%. m.p. 178–
181 °C (decomp.). KM (DMF solution, S cm2 molꢁ1): 0.3. ESI+ MS
(methanol, m/z): [Pd(ox)(L2)3 + H]+ 1189.4, [Pd(ox)(L2)2 + H]+
858.7, [Pd(ox)(L2) + H]+ 527.8, [L2 + H]+ 332.2. IR (Nujol; cmꢁ1):
m(C@C)ar; 1376vs
m
m
m
m
m
m
m
m
m
m
558vs
m
(Pd–O), 506s
m
(Pd–N). IR (KBr; cmꢁ1): 3114w, 3053w
(C–H)al; 1711s, 1675s (C@O)ox
(C@C)ar; 1373w (C–O)ox; 1265s
(Pd–O). Raman (cmꢁ1): 3370w
(C–H)ar; 2990s, 2939s, 2838w (C–H)al;
(C@O)ox; 1537w, 1485s (C@C)ar; 1167w (C–Cl);
(br, N6H, 1H), 8.38 (s, C8H, 1H), 7.50 (dd, 7.2, 1.6, C11H, C15H, 2H),
7.31 (tt, 7.2, 1.6, C12H, C14H, 2H), 7.23 (tt, 7.2, 1.6, C13H, 1H), 6.32
(br, N2H, 1H), 4.80 (d, 6.0, C9H, 2H), 4.70 (m, O20H, 1H), 4.67 (sp,
6.8, C16H, 1H), 3.97 (sx, 6.8, C19H, 1H), 3.66 (sx, 6.8, C20Ha, 1H),
3.59 (sx, C20Hb, 1H), 1.75 (sp, 6.8, C21Ha, 1H), 1.56 (m, C21Hb, 1H),
1.50 (d, 6.8, C17H, C18H, 6H), 0.93 (t, 6.8, C22H, 3H). 13C NMR
(DMF-d7, ppm): d 166.20 (C25, C26), 160.44 (C2), 153.50 (C6),
151.43 (C4), 140.57 (C10), 139.24 (C8), 128.80 (C12, C14), 128.32
(C11, C15), 127.21 (C13), 111.84 (C5), 64.12 (C20), 55.35 (C19), 48.39
(C16), 44.61 (C9), 24.76 (C21), 21.84 (C17), 21.80 (C18), 10.93 (C22).
15N NMR (DMF-d7, ppm): d 199.8 (N1), 181.0 (N9), 144.3 (N7),
96.7 (N2), 91.6 (N6).
m
(C–H)ar, 2977w, 2938w, 2836w
m
m
m
m
;
1618vs
m
(C@N); 1537w, 1488s
(C–Cl); 559w
(N–H); 3139w, 3059w
m
m
m
(C–O)ar; 1165w
m
m
m
1703w, 1659s
m
m
m
559s
m
(Pd–O). 1H NMR (DMF-d7, ppm): d 9.22 (t, 6.2, N6H, 1H),
8.77 (s, C8H, 1H), 7.23 (t, 8.0, C14H, 1H), 7.10 (t, 2.2, C11H, 1H),
7.09 (d, 7.8, C15H, 1H), 6.84 (dd, 8.2, 2.6, C13H, 1H), 4.85 (d, 6.2,
C9H, 2H), 4.81 (sp, 6.8, C16H, 1H), 3.82 (s, C23H, 3H), 1.53 (d, 6.8,
C
17H, C18H, 6H). 13C NMR (DMF-d7, ppm): d 165.72 (C25, C26),
160.48 (C12), 155.13 (C2), 154.05 (C6), 150.32 (C4), 143.81 (C8),
140.85 (C10), 130.04 (C14), 120.62 (C15), 117.26 (C5), 113.91 (C11),
113.31 (C13), 55.50 (C23), 49.64 (C16), 45.19 (C9), 21.96 (C17, C18).
15N NMR (DMF-d7, ppm): d 227.6 (N1), 217.6 (N3), 181.5 (N9),
143.4 (N7), 97.2 (N6).
5: Yield: 240 mg (50%). Anal. Calc. for C42H56N12O8Pd: C, 52.4; H,
5.9; N, 17.4. Found: C, 51.9, H, 5.8; N, 17.6%. m.p. 159–160 °C (de-
comp.). KM (DMF solution, S cm2 molꢁ1): 3.0. ESI+ MS (methanol,
m/z): [Pd(ox)(L5)3 + H]+ 1347.2, [Pd(L5) + H]+ 491.1, [L5 + H]+
3: Yield: 610 mg (63%). Anal. Calc. for C36H40N10O8Cl2Pdꢀ3H2O:
C, 44.5; H, 4.8; N, 14.4. Found: C, 44.1; H, 4.7; N, 13.9%. m.p.
176–179 °C (decomp.). KM (DMF solution, S cm2 molꢁ1): 0.1. ESI+
385.3. IR (Nujol; cmꢁ1): 559vs
cmꢁ1): 3116w, 3072w
(C–H)al; 1708s, 1676s
m
(Pd–O); 527vs
(C–H)ar, 2964w, 2933w, 2875w, 2837w
(C@O)ox; 1609vs (C@N); 1541s, 1492s
m(Pd–N). IR (KBr;
m
m
m
m

1472
P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
m
m
m
m
(C@C)ar; 1371s
(Pd–O); 526w
(C–H)ar; 2974s, 2936vs, 2878s, 2842w
m
m
(C–O)ox; 1243s
(Pd–N). Raman (cmꢁ1): 3345w
(C–H)al; 1705w, 1672w,
(C–N); 1536w, 1491s (C@C)ar; 1251s (C–
(Pd–O); 526w
(Pd–N). 1H NMR
m
(C–O)ar; 1050w
m
(C–O)al; 556w
2.3. Physical measurements
m(N–H); 3069s
m
Elemental analyses (C, H, N) were performed on a Fisons EA-
1108 CHNS-O Elemental Analyzer (Thermo Scientific). The yields
were calculated and based on palladium. Melting point determina-
tions were performed on a Melting Point B-540 apparatus (Büchi)
with 5 °C minꢁ1 gradient and the obtained values were uncor-
rected. Conductivity measurements were carried out on a Cond
340i/SET (WTW) in N,N0-dimethylformamide (DMF; 10ꢁ3 M) solu-
tion at the temperature of 25 °C. Infrared spectra were recorded
(C@O)ox; 1606vs
m
m
m
O)ar; 1049
m(C–O)al; 560w
m
m
(DMF-d7, ppm): d 8.34 (s, C8H, 1H), 8.23 (br, N6H, 1H), 7.40 (d,
7.5, C15H, 1H), 7.24 (tt, 7.9, 1.6, C13H, 1H), 7.02 (d, 8.2, C12H, 1H),
6.85 (t, 7.5, C14H, 1H), 6.29 (br, N2H, 1H), 4.80 (d, 7.3, C9H, 2H),
4.70 (br, O20H, 1H), 4.68 (sp, 6.8, C16H, 1H), 3.94 (m, C19H, 1H),
3.91 (s, C23H, 3H), 3.64 (m, C20Ha, 1H), 3.56 (m, C20Hb, 1H), 1.72
(sp, 7.4, C21Ha, 1H), 1.55 (sp, 7.4, C21Hb, 1H), 1.50 (d, 6.8, C17H,
on a Nexus 670 FT-IR (ThermoNicolet) by KBr (400–4000 cmꢁ1
)
C
18H, 6H), 0.91 (br, C22H, 3H). 13C NMR (DMF-d7, ppm): d 165.94
and Nujol (150–600 cmꢁ1) techniques. Raman spectroscopy was
performed on an NXR FT-Raman Module (ThermoNicolet) in the
150–3750 cmꢁ1 region; the Raman spectrum was not obtained in
case of 1 (the sample burnt under laser beam). The reported IR
and Raman signal intensities have been defined as w = weak,
s = strong and vs = very strong. 1H and 13C spectra and 1H–1H gs-
COSY, 1H–13C gs-HMQC, 1H–13C gs-HMBC and 1H–15N gs-HMBC
(C25, C26), 160.49 (C2), 157.95 (C11), 153.64 (C6), 151.40 (C4),
139.28 (C8), 128.87 (C13), 128.61 (C15), 127.90 (C10), 120.87 (C14),
112.07 (C5), 110.92 (C12), 64.05 (C20), 55.83 (C23), 55.38 (C19),
48.36 (C16), 39.75 (C9), 24.78 (C21), 21.86 (C17), 21.82 (C18), 10.96
(C22). 15N NMR (DMF-d7, ppm): d 200.0 (N1), 181.8 (N9), 180.0
(N3), 145.4 (N7), 96.9 (N2), 89.7 (N6).
6: Yield: 370 mg (77%). Anal. Calc. for C42H56N12O8PdꢀH2O: C,
51.4; H, 6.0; N, 17.1. Found: C, 51.0, H, 6.1; N, 17.2%. m.p. 155–
158 °C (decomp.). KM (DMF solution, S cm2 molꢁ1): 2.4. ESI+ MS
(methanol, m/z): [Pd(ox)(L6)3 + H]+ 1347.0, [Pd(L6) + H]+ 491.2,
(gs = gradient
HMQC = Heteronuclear
HMBC = Heteronuclear Multiple Bond Coherence) correlation
selected,
COSY = correlation
Multiple Quantum
spectroscopy,
Coherence,
experiments (DMF-d7 solutions of L1–7 and 1–7) were measured
[L6 + H]+ 385.3. IR (Nujol; cmꢁ1): 557vs
m
(Pd–O); 524vs
(C–H)ar, 2965w, 2934w, 2875w, 2834w
(C@O)ox; 1609vs (C@N); 1544s, 1491s
(C–O)ar; 1047w (C–O)al; 558w
(C–H)ar; 2975s, 2937vs, 2876s,
(C@O)ox 1608vs (C–N);
(C–O)ar; 559s (Pd–O); 526w
m
(Pd–N).
on
a
Varian 400 MHz NMR device at 400.00 MHz (1H),
IR (KBr; cmꢁ1): 3120w
m
m
100.58 MHz (13C) and 40.53 MHz (15N). Spectra were obtained at
natural abundance at 300 K (1H and 1H–15N gs-HMBC also at
340 K) and were calibrated against the signals of tetramethylsilane
(an internal standard for 1H and 13C NMR spectra) and against the
residual signals of the solvent (an internal reference for 15N ad-
justed to 104.7 ppm). The splitting of proton resonances in the re-
ported 1H spectra is defined as s = singlet, d = doublet, t = triplet,
sx = sextuplet, sp = septuplet, br = broad band, dd = doublet of dou-
blets, tt = triplet of triplets, m = multiplet. Mass spectra (MS) of the
methanol solutions of 1–7 were obtained using a LCQ Fleet ion trap
mass spectrometer by the positive mode electrospray ionization
(ESI+) technique (Thermo Scientific). Simultaneous thermogravi-
metric (TG) and differential thermal (DTA) analyses were carried
out using a thermal analyzer Exstar TG/DTA 6200 (Seiko Instru-
ments Inc.). TG/DTA studies were performed in ceramic pans from
laboratory temperature to 900 °C with a 2.5 °C minꢁ1 temperature
gradient in dynamic air atmosphere (100 mL minꢁ1). Geometry of
the complex 2 was fully optimized at the B3LYP level with the 6-
m
m
m
(C–H)al; 1708s, 1676s
(C@C)ar; 1374s
(Pd–O). Raman (cmꢁ1): 3061w
m
m
(C–O)ox; 1265s
m
m
m
2833w
m(C–H)al; 1701w, 1669w,
m
;
m
1542w, 1488w
m
(C@C)ar; 1266s
m
m
m
(Pd–N). 1H NMR (DMF-d7, ppm): d 8.46 (br, N6H, 1H), 8.36 (s,
C8H, 1H), 7.21 (t, 7.9, C14H, 1H), 7.13 (t, 2.1, C11H, 1H), 7.08 (d,
7.6, C15H, 1H), 6.81 (dd, 8.2, 2.6, C13H, 1H), 6.34 (br, N2H, 1H),
4.82 (br, O20H, 1H), 4.76 (d, 6.8, C9H, 2H), 4.68 (sp, 6.8, C16H, 1H),
3.95 (sx, 5.5, C19H, 1H), 3.80 (s, C23H, 3H), 3.66 (sp, 5.5, C20Ha,
1H), 3.55 (m, C20Hb, 1H), 1.74 (sp, 7.4, C21Ha, 1H), 1.55 (sp, 7.4,
C
21Hb, 1H), 1.50 (d, 6.8, C17H, C18H, 6H), 0.92 (t, 7.5, C22H, 3H).
13C NMR (DMF-d7, ppm): d 166.04 (C25, C26), 160.54 (C12), 160.44
(C2), 153.53 (C6), 151.46 (C4), 142.32 (C10), 139.32 (C8), 129.85
(C14), 120.60 (C15), 113.67 (C11), 113.11 (C13), 111.87 (C5), 64.12
(C20), 55.47 (C23), 55.36 (C19), 48.34 (C16), 44.68 (C9), 24.77 (C21),
21.88 (C17), 21.84 (C18), 10.96 (C22). 15N NMR (DMF-d7, ppm): d
199.6 (N1), 181.1 (N9), 144.7 (N7), 96.8 (N2), 91.7 (N6).
*
311G /LANL2DZ basis set, where the LANL2DZ pseudo-potential
7: Yield: 320 mg (66%). Anal. Calc. for C42H56N12O8Pd: C, 52.4; H,
5.9; N, 17.4. Found: C, 52.2, H, 6.4; N, 17.3%. mp 165–167 °C (de-
comp.). KM (DMF solution, S cm2 molꢁ1): 3.2. ESI+ MS (methanol,
m/z): [Pd(ox)(L7)3 + H]+ 1347.0, [Pd(L7) + H]+ 491.1, [L7 + H]+
was applied for the Pd(II) ion. Theoretical calculations were per-
formed with SPARTAN06 program package [13]. The molecular gra-
phic was drawn by DIAMOND
, the structural parameters and
calculations were interpreted using the same software [14].
385.3. IR (Nujol; cmꢁ1): 560vs
m
(Pd–O); 522vs
(C–H)ar, 2964s, 2933s, 2875s, 2835w
(C@O)ox; 1608vs (C@N); 1543vs, 1493s
(C–O)ar; 1057w (C–O)al; 561w
(Pd–N). Raman (cmꢁ1): 3335w
(N–H); 3058s
H)ar; 2976s, 2936vs, 2876s, 2842w (C–H)al; 1711w, 1672w,
(C@O)ox 1610vs (C–N); 1549w (C@C)ar; 1255w (C–O)ar;
561w
(Pd–O). 1H NMR (DMF-d7, ppm): d 8.36 (br, N6H, 1H), 8.33
(s, C8H, 1H), 7.45 (dd, 8.6, 2.0, C12H, C14H, 2H), 6.87 (dd, 8.6, 2.0,
m
(Pd–N). IR (KBr;
(C–H)al;
(C@C)ar;
(Pd–
(C–
cmꢁ1): 3132w
1707vs, 1675s
m
m
m
m
2.4. In vitro cytotoxic activity
m
1374vs
m
(C–O)ox; 1249vs
m
m
m
In vitro cytotoxicity of the complexes 1, 3–5 and 7 was evalu-
ated by an MTT assay against the human osteosarcoma cancer cell
line (HOS); [MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltet-
razolium bromide] [15].
O); 525w
m
m
m
m
m
;
m
m
m
m
The suspension of 2.5 ꢂ 104 cells/well was stabilized in 96-well
microplates in the culture medium enriched by fetal calf serum for
16 h at the temperature of 37 °C in 5% CO2 atmosphere. The tested
C
11H, C15H, 2H), 6.35 (br, N2H, 1H), 4.77 (t, 6.0, O20H, 1H), 4.71
(d, 5.5, C9H, 2H), 4.67 (sp, 6.6, C16H, 1H), 3.99 (br, C19H, 1H), 3.78
(s, C23H, 3H), 3.69 (m, C20Ha, 1H), 3.57 (m, C20Hb, 1H), 1.77 (sp,
6.8, 1.5, C21Ha, 1H), 1.58 (sp, 6.8, C21Hb, 1H), 1.49 (d, 6.8, C17H,
complexes were dissolved in DMF up to concentration of 50 lM,
and then diluted with the cell culture medium to the final DMF
concentration of 0.1%. This mixture was added to microplates with
the cancer cells instead of the above-mentioned culture medium.
The cancer cells were incubated for the period of 24 h. After this
period, the cells were washed with sterile phosphate buffer saline
C
18H, 6H), 0.94 (tt, 7.5, 2.5, C22H, 3H). 13C NMR (DMF-d7, ppm): d
166.10 (C25, C26), 160.48 (C2), 159.31 (C13), 153.48 (C6), 151.42
(C4), 139.30 (C8), 132.45 (C10), 129.83 (C12, C14), 114.24 (C11, C15),
111.87 (C5), 64.19 (C20), 55.48 (C23), 55.40 (C19), 48.42 (C16),
44.14 (C9), 24.81 (C21), 21.81 (C17,18), 10.97 (C22). 15N NMR
(DMF-d7, ppm): d 200.0 (N1), 181.2 (N9), 144.4 (N7), 97.3 (N2),
93.3 (N6).
(PBS), and 100
l
L of MTT (0.3 mg mLꢁ1) were poured in. The med-
L of
ium including the complexes was removed after 2 h. 100
l
DMSO with 1% NH3 were added to dissolve the purple formazane,
whose absorbance was measured at 630 nm (an automatic

P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
1473
microplate ELISA reader). The obtained results are discussed as IC50
values.
An oxalate dianion, which is bidentate-coordinated to the Pd(II)
ion in complexes 1–7, showed two bands of the as(C@O)ox vibra-
tion at 1671–1676 and 1706–1712 cmꢁ1 [20]. Other peaks of an
m
oxalato group, assignable to the ms(C–O)ox vibration, were found
in the 1364–1378 cmꢁ1 region. The coordination of both ligand
types to the metal centre was indirectly proved by detection of
3. Results and discussion
the
of the
m(Pd–N) and m(Pd–O) vibrations in the far-IR spectra. The bands
m
3.1. General features
(Pd–O) were observed at 557–560 cmꢁ1, while the maxima
observed between 506 and 527 cmꢁ1 are assignable to the
vibration [19–21].
m(Pd–N)
The palladium(II) oxalato complexes of the general composition
[Pd(ox)(L1–7)2]ꢀnH2O (1–7; n = 0 for 4, 5 and 7, 3 for 1 and 2, 1 for 6
4
Most of the above-described vibrations detected by the IR spec-
troscopy of 1–7 were observed in the corresponding Raman spectra
of the complexes 2–7 (complex 1 burnt up in the light of laser
and 3 for 3) have been prepared by a one-step synthesis using the
reaction of K2[Pd(ox)2]ꢀ2H2O with two molar equivalents of the
appropriate N6-(benzyl)-9-isopropyladenine-based organic mole-
cule (L1–7; specified in Section 2.1 and Scheme 1). The pale yellow
powder complexes 1–7 were isolated after two days of stirring at
40 °C by filtration from the reaction mixture (Scheme 2).
beam). The maxima belonging to the
(Pd–O) stretching vibrations were found in Raman spectra of all
the complexes 2–7 at 3057–3148, 2833–2990, and at 559–
561 cmꢁ1, respectively. Two maxima of the
as(C@O)ox vibrations
were detected at 1692–1711 and 1657–1672 cmꢁ1, while the max-
ima of the (C@C)ar were observed at 1535–1549 and 1454–
1485 cmꢁ1. However, the other vibrations discussed for IR spec-
troscopy, i.e. (C@N)ar, s(C–O)ox (C–O)ar and (Pd–N), were ob-
served only in Raman spectra of some of the palladium(II)
oxalato complexes 2–7 (see Section 2.2). In case of the (C@N)ar
m(C–H)ar, m(C–H)al, and
m
m
The complexes 1–7 were determined to be non-electrolytes,
since the molar conductivity values of their 10ꢁ3 M DMF solutions
ranged from 0.1 to 5.8 S cm2 molꢁ1, and these values are typical for
non-electrolytes in DMF [16].
m
m
m
,
m
m
The prepared palladium(II) oxalato complexes 1–7 are well sol-
uble in various solvents (e.g. DMF, DMSO, chloroform, ethanol,
methanol, acetone) and less soluble in water. However, the at-
tempts to prepare crystals suitable for a single-crystal X-ray anal-
ysis have been unsuccessful to date in all the above-mentioned
solvents or their combinations. That is why the composition and
structure of complexes 1–7 were deduced from the obtained re-
sults of physical techniques (mainly from NMR and mass spectra)
and based on similarity with the previously published [Pd(ox)(LI)2]
(I) and [Pd(ox)(LII)2]ꢀLIIꢀMe2CO (II) complexes whose structures
were determined by a single-crystal X-ray analysis; LI = 2-chloro-
N6-(4-methoxybenzyl)-9-isopropyladenine and LII = 2-chloro-N6-
(4-methylbenzyl)-9-isopropyladenine [3]. For the lack of single
crystals suitable for a crystallographic study, the geometry of the
complex 2 was optimized using DFT calculations (see Section 3.5).
m
vibration, it could be connected with decrease in intensity of the
named band, as recently experimentally observed and theoretically
calculated for this vibration in adenine [18]. Nevertheless, in the
cases that these bands were observed, the positions of their max-
ima correlated well with those observed in the IR spectra (see Sec-
tion 2.2 for more details). The band of a very strong intensity at
about 1320 cmꢁ1, which was not detected in the IR spectra, can
be assigned to the skeletal stretching vibrations of a purine ring,
as formerly reported [18,22]. Contrary to the IR spectra, the m(N–
H) bands were detected in the Raman spectra of 2, 3, 5 and 7
(not for 4 and 6). However, the maxima of the peaks connected
with this vibration, observed within the 3333–3370 cmꢁ1 region,
were of very low intensity, which is in accordance with results gi-
ven for non-substituted adenine [18].
3.2. IR and Raman spectroscopy
3.3. NMR spectroscopy
The title [Pd(ox)(L1–7)2]ꢀnH2O complexes (1–7) involve two
types of ligands, i.e. the N-donor ligands (L1–7) derived from N6-
(benzyl)-9-isopropyladenine, and O-donor bidentate-coordinated
oxalate dianion (ox). Characteristic vibrations of both ligand types
were unambiguously detected in the IR (150–4000 cmꢁ1) and Ra-
man (150–3750 cmꢁ1) spectra.
Multinuclear and two dimensional (2D) NMR experiments (see
Section 2.3) were performed for all the complexes 1–7 as well as
for the free organic compounds L1–7 involved in 1–7 as N-donor li-
gands. The chemical shifts (d; ppm) are given in Section 2.2, while
the coordination shifts (
rized in Tables 1–3.
D
d = dcomplex ꢁ dligand; ppm) are summa-
The L1–7 organic molecules coordinated to the central Pd(II) ion
in the complexes 1–7 showed characteristic and the most intensive
All the signals detected in the 1H and 13C NMR spectra of the
free L1–7 molecules were also found in the appropriate spectra of
m
(C@N) bands at 1608–1622 cmꢁ1 [17,18]. It should be pointed
out, that the maxima values differ between the groups of com-
plexes 1–3 (1618–1622 cmꢁ1) and 4–7 (1608–1610 cmꢁ1), which
may be connected with a different substitution of N6-(benzyl)-9-
isopropyladenine skeleton at the C2 position. The peaks of the
complexes 1–7. The highest |D
d| values in the 1H NMR spectra
were observed for the hydrogen atoms bound to the N6 (0.34–
0.95 ppm downfield) and C8 (0.28–0.61 ppm downfield) atoms
(see Table 1). In the case of complexes 4–7, the N2H signals are sig-
nificantly shifted downfield as well, but less than the mentioned
N6H and C8H signals. The coordination shifts of the other proton
signals were insignificant.
m
(C–H)ar and m(C–H)al vibrations were detected at 3030–3137
and 2834–2982 cmꢁ1, respectively [17–19]. The
m(C@C)ar bands
showed two maxima at ꢃ1470 and ꢃ1545 cmꢁ1. The bands ob-
served in the 1229–1265 cmꢁ1 region are assignable to the
m
(C–
The highest coordination shifts of carbon atoms were deter-
mined for the C8 atom (3.50–3.94 ppm downfield), followed by
the C5 (1.24–3.12 ppm upfield), C6 (0.93–2.22 ppm upfield) and
C16 (1.31–1.75 ppm downfield) ones (Table 2). It can be pointed
out, that the C5, C6 and C8 coordination shifts of 1–3 and 4–7 differ
significantly within the group of complexes 1–7, which is most
likely caused by various compositions and structures of the appro-
priate adenine derivatives involved in these two types of com-
pounds. It has to be noted, that there was one more signal
detected at ꢃ166 ppm in the 13C NMR spectra of the complexes
1–7, which was not observed in the spectra of free L1–7 molecules,
O)ar vibration of the methoxy substituent of a benzyl group (it
was not found in case of 4, which does not have the methoxy-
substituted benzene ring). The presence of the m(C–Cl) (for 1–3)
and
m
(C–O)al (for 4–7) vibrations at ꢃ1050 and ꢃ1160 cmꢁ1
,
respectively, is a consequence of a different substitution of the
C2 atom of the adenine derivatives L1–3 and L4–7 involved in the
complexes 1–3 and 4–7. A broad peak was observed in IR spectra
of all the complexes between ca. 3300 and 3500 cmꢁ1, and thus,
the maxima belonging to both the
m(N–H) and m(O–H) vibrations
could not be assigned unambiguously.

Table 1
Coordination shifts (
D
d = dcomplex ꢁ dligand) determined from the 1H NMR spectra for the complexes 1–7.
N2H
N6H
C8H
C9H
C11
H
C12
H
C13
H
C14
H
C15
H
C16
H
C17H, C18
H
C19
H
C20Ha
C20Hb
O20
H
C21Ha
C21Hb
C22
H
C23
H
C24
H
1
2
3
4
5
6
7
0.61
0.34
0.52
0.71
0.95
0.77
0.76
0.40
0.28
0.47
0.61
0.54
0.56
0.55
0.08
0.05
0.07
ꢁ0.03
0.02
0.01
0.01
0.02
ꢁ0.03
0.00
0.03
ꢁ0.03
0.12
0.07
0.07
0.05
0.06
0.06
ꢁ0.01
0.05
0.01
0.03
0.06
0.06
0.07
0.06
ꢁ0.05
ꢁ0.06
ꢁ0.04
ꢁ0.02
ꢁ0.03
ꢁ0.04
ꢁ0.03
0.04
0.03
0.03
0.02
0.07
0.05
ꢁ0.01
0.03
0.46
0.49
0.48
0.49
0.01
0.00
0.01
0.01
ꢁ0.01
ꢁ0.01
ꢁ0.02
0.01
0.00
0.00
ꢁ0.07
0.00
ꢁ0.09
ꢁ0.07
0.06
ꢁ0.01
0.01
0.00
0.00
0.02
ꢁ0.18
ꢁ0.04
ꢁ0.03
0.00
ꢁ0.01
ꢁ0.01
0.01
ꢁ0.01
0.01
0.00
ꢁ0.02
ꢁ0.02
0.07
0.01
0.03
0.01
0.06
ꢁ0.01
ꢁ0.02
ꢁ0.03
0.07
0.02
0.00
Table 2
Coordination shifts (
D
d = dcomplex ꢁ dligand) determined from the 13C NMR spectra for the complexes 1–7.
C2
C4
C5
C6
C8
C9
C10
C11
C12
C13
C14
C15
C16
C17
C18
C19
C20
C21
C22
C23
C24
C25,26
1
2
3
4
5
6
7
0.24
0.80
0.19
0.13
0.21
0.18
0.20
ꢁ0.21
ꢁ0.06
ꢁ0.21
ꢁ0.42
ꢁ0.28
ꢁ0.34
ꢁ0.30
ꢁ2.44
ꢁ1.24
ꢁ2.37
ꢁ3.12
ꢁ2.97
ꢁ3.09
ꢁ3.08
ꢁ1.09
ꢁ1.60
ꢁ0.93
ꢁ2.20
ꢁ2.22
ꢁ2.12
ꢁ2.14
3.77
3.94
3.70
3.50
3.53
3.58
3.63
0.92
0.95
1.08
0.68
0.67
0.93
0.95
ꢁ0.94
ꢁ0.86
ꢁ1.03
ꢁ1.15
ꢁ0.97
ꢁ1.06
ꢁ1.18
0.05
0.08
ꢁ0.03
0.01
0.02
0.04
0.13
0.23
0.24
0.43
0.70
0.01
0.36
0.61
0.06
0.18
ꢁ0.01
0.01
0.02
0.17
0.06
0.23
0.58
0.24
0.11
0.08
0.16
0.17
0.07
1.75
1.31
1.74
1.55
1.52
1.54
1.63
ꢁ0.46
ꢁ0.37
ꢁ0.41
ꢁ0.54
ꢁ0.51
ꢁ0.48
ꢁ0.50
ꢁ0.46
ꢁ0.37
ꢁ0.41
ꢁ0.54
ꢁ0.51
ꢁ0.49
ꢁ0.46
0.16
0.06
0.12
ꢁ1.38
ꢁ1.23
ꢁ1.36
ꢁ0.75
ꢁ1.01
ꢁ0.91
ꢁ0.85
ꢁ0.23
0.11
0.12
0.08
0.07
0.02
0.12
0.06
0.09
ꢁ0.44
ꢁ0.19
ꢁ0.10
ꢁ0.39
ꢁ0.39
ꢁ0.38
ꢁ0.13
ꢁ0.15
ꢁ0.13
ꢁ0.03
ꢁ0.05
ꢁ0.06
0.12
0.11
0.02
ꢁ0.36
0.07

P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
1475
Table 3
15N coordination shifts (
D
d = dcomplex ꢁ dligand) determined from the 1H–15N gs-HMBC
NMR spectra for the complexes 1–7.
N1
N2
N3
N6
N7
N9
1
2
3
4
5
6
7
2.7
1.5
3.6
2.0
ꢁ1.3
ꢁ0.3
ꢁ2.8
ꢁ0.9
ꢁ5.5
ꢁ0.9
n.o.
8.2
1.6
6.5
4.0
5.7
8.8
12.6
ꢁ93.0
ꢁ98.2
ꢁ92.7
ꢁ96.5
ꢁ99.0
ꢁ94.5
ꢁ102.6
7.3
1.9
6.4
6.7
3.7
7.3
1.5
2.6
ꢁ1.3
ꢁ4.5
3.6
n.o.
ꢁ3.4
n.o.
n.o. – signals were not observed in the 1H–15N gs-HMBC spectra of the appropriate
complexes.
as well as in any of 2D carbon experiments. This signal unambigu-
ously belongs to the C25 and C26 atoms of an oxalate dianion
bidentate-coordinated to the metal centre. The signals of these
atoms are shifted by 0.75–1.38 ppm upfield as compared with
the corresponding 13C NMR signal of the starting compound
K2[Pd(ox)2]ꢀ2H2O dissolved in D2O and detected at 166.95 ppm.
Fig. 1. Observed (full line) and calculated (dashed line) isotopic distribution
representation of [Pd(ox)(L1) + H]+ (left) and [Pd(ox)(L1)2 + K]+ (right) peaks as
obtained by an ESI+ mass spectrometry of the complex 1.
As in the case of the C5, C6 and C8 atoms, the | d| of C25 and C26
D
cept for [Pd(L4)2 + H]+ observed at 813.1 m/z; the [Pd(ox)(L4)2 + -
Na]+ and [Pd(ox)(L4) + Na]+ adducts were detected for this
compound as well. On the other hand, the [Pd(L4–7) + H]+ fragment,
whose analogue was not detected in any spectrum of 1–3, was
clearly observed for the compounds 4–7 m/z at 461.2, 491.1,
491.2, and 491.1, respectively. Similarly as in the case of complexes
1–3, the [Pd(ox)(L4–7)3 + H]+ and [(L4–7) + H]+ peaks were found in
the mass spectra of 4–7.
are different for the complexes 1–3 as compared with 4–7.
In the case of 1H–15N gs-HMBC 2D correlation experiments,
coordination shifts of the N7 atoms [
D
d = ꢁ92.7 ꢁ (ꢁ102.6) ppm]
were found to be significantly higher as compared with those of
the remaining nitrogen atoms (| d| < 12.6 ppm) involved in L1–7
D
molecules (Table 3). The analogical NMR spectroscopy results were
recently published for palladium(II) oxalato [3] and dichlorido [5]
complexes involving N6-(benzyl)-9-isopropyladenine-based N-do-
nor ligands, whose structures were crystallographically deter-
mined. It is caused by a coordination of L1–7 molecules to the
Pd(II) ion through the mentioned N7 atom of the adenine moiety
within the structure of the complexes 1–7. In connection with this
statement, we can focus back on the results of 1H and 13C NMR
spectroscopy, where we discuss the C8H, N6H, C8 and C5 atoms as
3.5. DFT calculations
The molecular structures of [Pd(ox)(LI)2] (I) and [Pd(ox)(-
LII)2]ꢀLIIꢀMe2CO (II) complexes were determined by a single-crystal
X-ray analysis, as it was reported in our previous work describing
the palladium(II) oxalato complexes with N6-(benzyl)-9-isopropy-
those with the highest
Dd. As it can be seen from Schemes 1 and
2, the above-mentioned hydrogen and carbon atoms neighbour
to the coordination site, i.e. the N7 atom, which indirectly support
discussed conclusion regarding coordination.
3.4. ESI+ mass spectrometry
The ESI+ MS of the complexes 1–7 dissolved in methanol were
measured and the results are given in Section 2.2. All the observed
isotopic distribution representations correspond very well with the
theoretic ones (QUALBROWSER software, version 2.0.7, Thermo Fischer
Scientific).
In case of the complexes 1–3, involving 2-chloro-N6-(benzyl)-9-
isopropyladenine-based molecules L1–3, the [Pd(ox)(L1–3)2 + H]+
molecular peaks were found in the appropriate spectra at
858.8 m/z (1), 858.7 m/z (2) and 918.6 m/z (3), which indirectly
confirmed the composition of discussed palladium(II) oxalato com-
plexes. Moreover, the [Pd(ox)(L1)2 + Na]+ (for 1), [Pd(ox)(L1)2 + K]+
(for 1) and [Pd(ox)(L1–3)3 + H]+ (for 1–3) adducts were observed
as well. The [Pd(ox)(L1–3) + H]+ and [(L1–3) + H]+ fragments of the
studied compounds were also detected in the mass spectra of 1–
3; [(L1–3) + H]+ fragment, detected at 332.1 m/z for 1, 332.2 m/z
for 2 and 362.1 m/z for 3, is unambiguously assignable to the ade-
nine derivative (L1–3) involved in the structures of 1–3. Fig. 1 de-
picts the [Pd(ox)(L1) + H]+ and [Pd(ox)(L1)2 + K]+ peaks found in
ESI+ mass spectrum of the complex 1 and their comparison with
the theoretic ones.
Quite different results were obtained for the complexes 4–7,
since any of the [Pd(ox)(L4–7)2 + H]+, [Pd(L5–7)2 + H]+, or
[Pd(ox)(L4–7) + H]+ peaks were not found in their mass spectra, ex-
*
Fig. 2. Geometry of the [Pd(ox)(L2)2] (2) complex optimized on the B3LYP/6-311G /
LANL2DZ level of theory.

1476
P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
ladenine-based N-donor ligands; LI = 2-chloro-N6-(4-methoxyben-
zyl)-9-isopropyladenine, LII = 2-chloro-N6-(4-methylbenzyl)-9-
significantly from those of I and II. But differences of the same or-
der can be found even between some angles of I and II [e.g.
O(2)ꢁPd(1)ꢁN(7A) angle], which is caused by a variedness of the
structures of the studied complexes (see Table 4).
isopropyladenine [3]. In present work, partially due to a lack of
single crystals suitable for a crystallographic study, we decided
to perform theoretical calculations relating to the geometry of
*
the complex [Pd(ox)(L2)2] (2) on the B3LYP/6-311G /LANL2DZ le-
3.6. Thermal analysis
vel, whose optimized structure is depicted in Fig. 2. The selected
bond lengths and angles of the complexes 2, I and II are given in
Table 4.
The TG and DTA methods of a thermal analysis of 1–7 indicated
two types of thermal behaviour of the complexes 1–3 and 4–7. As
representatives of both groups, the complexes 3 and 4 were chosen
for detailed interpretation and their TG/DTA curves are depicted in
Fig. 3. All the important thermal characteristics of the studied com-
plexes are given in Table 5.
The central Pd(II) ion of the complex 2 is tetra-coordinated by
two 2-chloro-N6-(3-methoxybenzyl)-9-isopropyladenine (L2) mol-
ecules bound to the metal centre through their N(7) atoms of the
adenine moieties and by one bidentate-coordinated oxalate dian-
ion. Geometry in the vicinity of the central Pd(II) ion is distorted
square-planar. The deviations from a least-square plane defined
by atoms of the PdN2O2 chromophore are: 0.027 Å for Pd(1),
0.070 Å for O(1), ꢁ0.089 Å for O(2), 0.064 Å for N(7) and ꢁ0.072 Å
for N(7A). For the complex II, the above-discussed deviations were
determined to be: 0.0005(3) Å [Pd(1)], ꢁ0.0305(22) Å [O(1)],
0.0276(27) Å [O(2)], ꢁ0.0513(29) Å [N(7)] and 0.0259(29) Å
[N(7A)]. In case of the complex II they equalled, in the given
order, 0.0032(3), ꢁ0.0202(23), ꢁ0.0643(23), ꢁ0.0530(32), and
ꢁ0.1318(32) Å, respectively. Two L2 molecules are, similarly to
the complex II, mutually arranged in the head-to-tail arrangement
within the structure of the complex 2; the complex I has its LI mol-
ecules arranged in the head-to-head orientation.
The dehydration of 3 began right after the start of the analysis at
28 °C, and this first step is finished at 132 °C. The endothermic ef-
fects (endo-effects), anticipated for the dehydration process, were
found on the DTA curve with minima at 40 and 124 °C. The com-
pound existed in its dehydrated form of [Pd(ox)(L3)2] between
132 and 149 °C. Consequently, the decay proceeded in three waves
without formation of thermally stable intermediates up to 480 °C.
This process is accompanied by two exothermic effects (exo-ef-
fects) on the DTA curve (Table 5) with maxima at 182, and
368 °C, respectively. A plateau, which may be connected with the
formation of a thermally stable intermediate palladium(II) oxide
(PdO), occurred on the TG curve from 480 up to 808 °C. The PdO
further decomposed to Pd (with the minimum of endo-effect at
816 °C), which formed as a final product of the thermal degrada-
tion of 3. No weight changes were observed on the TG curve from
820 °C to the final temperature of the experiment (900 °C). The ob-
served weight losses of described partial processes of the thermal
decomposition of 3 differ insignificantly from the calculated ones
(Table 5).
All the calculated Pd–N and Pd–O bands of 2 are slightly longer
as compared with those determined for the complexes I and II (Ta-
ble 4). As for bond angles, it has to be noted that in some cases [e.g.
O(1)ꢁPd(1)ꢁN(7)], the values observed for the complex 2 differ
Table 4
The anhydrous complex [Pd(ox)(L3)2] (4) is thermally stable up
to 128 °C (Fig. 3). The process of thermal degradation proceeded
from this temperature, with three steps observable on the TG
curve, without formation of any thermally stable intermediates
up to 544 °C, when it is finished by a formation of PdO. As it is gi-
ven in Table 5, four exo-effects and two endo-effects were detected
on the DTA curve in connection with this process. The second weak
endo-effect, whose minimum is lying at 163 °C, is most likely con-
nected with melting and simultaneous decomposition of the com-
plex 4 (melting temperature determined by a melting point
apparatus was found to be 164–166 °C). The PdO was thermally
stable in the 544–803 °C range. Its decomposition to Pd was ob-
served at 803–837 °C and it was accompanied by endo-effect on
Selected bond lengths (Å) and angles (°) for the complex [Pd(ox)(L2)2] (2) optimized
on the B3LYP/6-311G /LANL2DZ level of theory, and for the complexes [Pd(ox)(-
*
LI)2] (I) and [Pd(ox)(LII)2]ꢀLIIꢀMe2CO (II) as determined by
a single-crystal X-ray
analysis (see Ref. [3]); L2 = 2-chloro-N6-(3-methoxybenzyl)-9-isopropyladenine,
LI = 2-chloro-N6-(4-methoxy-benzyl)-9-isopropyladenine and LII = 2-chloro-N6-(4-
methylbenzyl)-9-isopropyladenine.
Compound
2
I
II
Bond lengths
Pd(1)ꢁN(7)
Pd(1)ꢁN(7A)
Pd(1)ꢁO(1)
Pd(1)ꢁO(2)
O(1)ꢁC(26)
O(2)ꢁC(25)
O(3)ꢁC(26)
O(4)ꢁC(25)
C(25)ꢁC(26)
2.078
2.110
2.017
2.011
1.355
1.351
1.234
1.233
1.540
2.012(3)
2.024(3)
1.992(2)
1.987(2)
1.302(4)
1.289(4)
1.215(4)
1.222(4)
1.549(5)
2.023(3)
2.021(3)
1.971(2)
1.983(2)
1.332(4)
1.282(4)
1.239(4)
1.230(4)
1.471(5)
Bond angles
O(1)ꢁPd(1)ꢁN(7A)
O(1)ꢁPd(1)ꢁN(7)
O(1)ꢁPd(1)ꢁO(2)
O(2)ꢁPd(1)ꢁN(7)
O(2)ꢁPd(1)ꢁN(7A)
N(7)ꢁPd(1)ꢁN(7A)
Pd(1)ꢁO(1)ꢁC(26)
Pd(1)ꢁO(2)ꢁC(25)
O(1)ꢁC(26)ꢁO(3)
O(2)ꢁC(25)ꢁO(4)
O(1)ꢁC(26)ꢁC(25)
O(2)ꢁC(25)ꢁC(26)
O(3)ꢁC(26)ꢁC(25)
O(4)ꢁC(25)ꢁC(26)
Pd(1)ꢁN(7)ꢁC(5)
Pd(1)ꢁN(7)ꢁC(8)
C(5)ꢁN(7)ꢁC(8)
95.37
168.73
82.28
86.83
173.54
95.76
114.48
114.67
122.62
123.38
114.02
114.33
123.36
122.30
134.37
118.18
106.45
134.31
119.50
106.11
93.09(10)
174.96(10)
84.46(9)
91.12(10)
177.10(10)
91.39(11)
111.8(2)
111.8(2)
125.2(3)
124.9(3)
115.3(3)
115.9(3)
119.5(3)
119.3(3)
127.9(2)
126.9(2)
105.2(3)
129.8(2)
124.8(2)
105.4(3)
89.86(10)
173.26(10)
84.20(10)
89.38(10)
171.75(10)
96.32(11)
110.6(2)
111.5(2)
121.1(3)
123.0(3)
116.3(3)
117.4(3)
122.5(3)
119.6(3)
135.8(2)
119.2(2)
105.0(3)
129.9(2)
119.6(2)
105.3(3)
Pd(1)ꢁN(7A)ꢁC(5A)
Pd(1)ꢁN(7A)ꢁC(8A)
C(5A)ꢁN(7A)ꢁC(8A)
Fig. 3. TG/DTA curves of the complexes 3 and 4 as obtained in the dynamic air
atmosphere in the 25–900 °C temperature range.

P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
1477
Table 5
Results of TG/DTA thermal analyses of the palladium(II) oxalato complexes 1–7.
Complex
Dehydration
[Pd(ox)(Ln)2]
[Pd(ox)(L1–7)2] ? PdO
PdO
PdO ? Pd
T (°C)a
DTA (°C)d
T (°C)a
D
m (%)b
T (°C)c
T (°C)a
D
m (%)b
T (°C)c
D
m (%)b
endo-effect
exo-effect
3
61–149
28–145
28–145
1.6/1.6
1.6/1.5
5.6/5.7
149–445
145–475
145–480
128–544
109–569
130–578
133–552
84.4/83.5
84.4/83.5
81.9/82.5
86.5/85.3
87.3/86.0
85.7/85.4
87.3/85.9
445–816
475–821
480–808
544–803
569–809
578–806
552–811
816–845
821–851
808–835
803–837
809–831
806–821
811–836
1.8/1.8
1.8/1.6
1.6/1.6
1.8/1.6
1.7/1.7
1.7/1.7
1.7/1.6
828
824
171, 416
178, 419
182, 368
170, 278, 396, 444
170, 301, 353, 407
173, 300, 429
[Pd(ox)(L1)2]ꢀ 4H2O (1)
3
[Pd(ox)(L2)2]ꢀ 4H2O (2)
[Pd(ox)(L3)2]ꢀ3H2O (3)
[Pd(ox)(L4)2] (4)
132–149
29–128
30–109
40, 124, 816
138, 163, 817
158, 823
156, 815
164, 825
[Pd(ox)(L5)2] (5)
[Pd(ox)(L6)2]ꢀH2O (6)
28–130
1.8/1.6
[Pd(ox)(L7)2] (7)
28–133
172, 287, 411
a
b
c
The temperature range of the corresponding transformation process.
Weight losses; calcd./found.
The temperature range of a thermally stable intermediate.
d
Positions of minima of endothermic effects (endo-effect) and maxima of exothermic effects (exo-effect).
the DTA curve with a minimum at 817 °C. A plateau of thermally
stable palladium can be seen above 837 °C to the final temperature
of 900 °C. Again, the obtained and calculated weight losses corre-
lated well with each other (see Table 5).
Based on the obtained results, the complexes 1–7 have been
characterized as square–planar compounds with the central Pd(II)
ion tetra-coordinated by one bidentate O-donor oxalate dianion
and two adenine derivatives bound to the metal centre through
N7 atoms of their adenine moieties, thus giving a PdN2O2 chromo-
phore. The prepared complexes 1, 3–5 and 7 were tested in vitro for
their antitumour activity against human osteosarcoma cancer cell
line, HOS. The obtained results showed the complexes 5 and 7 as
the substances with in vitro cytotoxicity comparable with commer-
cially used platinum-based drug Cisplatin.
3.7. In vitro cytotoxicity
Promising in vitro cytotoxicity (tested by an AM assay) results of
several palladium(II) oxalato complexes with 2-chloro-N6-(ben-
zyl)-9-isopropyladenine-based derivatives against K562 and
MCF-7 human cancer cell lines were reported in our previous pa-
per [3]. That is why we decided to test several representatives
(namely the complexes 1, 3–5 and 7) of a new series of palla-
dium(II) oxalato complexes involving mentioned N-donor ligands.
However, an MTT assay was used instead of an AM one, which re-
duces comparability of both groups of results.
Acknowledgements
The financial support from the Ministry of Education, Youth and
Sports of the Czech Republic is gratefully acknowledged (a Grant
No. MSM6198959218). The authors also thank Mrs. Pavla Richte-
rová for performing CHN elemental analyses, Ms. Radka Novotná
for infrared and Raman spectra measurements, Ms. Alena Klani-
cová for mass spectra measurements and Dr. Radim Vrzal and Prof.
The tested complexes showed the following in vitro cytotoxic
activities against HOS cancer cells: >50.0
l
M for 1, >5.0
lM for 3,
>25.0 M for 4, 34.9 11.0 M for 5, and 39.2 6.0 l
l
l
M for 7. The
obtained values were compared with that of platinum-based anti-
cancer drug Cisplatin, whose IC50 value was found to be
ˇ
ˇ
Zdenek Dvorák for in vitro cytotoxicity testing.
34.2 6.4 lM, as determined by an MTT test. It can be seen that
Appendix A. Supplementary material
in vitro cytotoxicity of tested complexes 5 and 7 is comparable with
that of Cisplatin. The results presented herein as well as in our pre-
vious paper [3] encourage us to continue with the biological test-
ing and to test these complexes by an MTT assay for their in vitro
cytotoxicity against some other human cancer cell lines, which is
currently in progress. Moreover, the DNA interaction study, as well
as the in vivo testing will be carried out for the palladium(II) oxa-
lato complexes involving N6-(benzyl)-9-isopropyladenine-based
derivatives in the near future.
A scheme of the synthetic pathway of L1–7 organic compounds,
as well as their IR, Raman and NMR (1H, 13C, 15N) spectral data, are
deposited. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.ica.2010.01.035.
References
[1] P.J. Davies, Plant Hormones, 3rd ed., Springer, Dordrecht, 1997.
[2] M. Gielen, E.R.T. Tiekink, Metallotherapeutic Drugs and Metal-based
Diagnostic Agents, Willey, London, 2005.
4. Conclusions
ˇ
[3] P. Štarha, Z. Trávnícek, I. Popa, J. Inorg. Biochem. 103 (2009) 978.
[4] A. Garoufis, S.K. Hadjikakou, N. Hadjiliadis, Coord. Chem. Rev. 253 (2008)
1384.
A series of [Pd(ox)(L1–7)2]ꢀnH2O (1–7; n = 0 for 4, 5 and 7, 3 4 for 1
and 2, 1 for 6 and 3 for 3) palladium(II) oxalato complexes was syn-
thesized by reactions of K2[Pd(ox)2]ꢀ2H2O with two molar equiva-
lents of the appropriate L1–7 organic compound. The pale yellow
powder products were isolated in good yields and characterized by
various physical methods. Within the complexes 1–7, compounds
1–3 represent analogues of recently published palladium(II)
oxalato complexes with 2-chloro-N6-(benzyl)-9-isopropyladenine
derivatives [3] similar to L1–3 molecules, but with differently
substituted benzyl group. On the other hand, we report here the
substances 4–7 as the first palladium(II) oxalato complexes involv-
ing the highly potent CDK inhibitors, namely 2-(1-ethyl-2-
hydroxyethylamnio)-N6-(benzyl)-9-isopropyladenine (Roscovitine;
L4) or its analogues with the substituted benzyl group (L5–7), as N-
donor carrier ligands.
ˇ
ˇ
[5] Z. Trávnícek, L. Szücová, I. Popa, J. Inorg. Biochem. 101 (2007) 477.
[6] K.I. Lee, T. Tashiro, M. Noji, Chem. Pharm. Bull. 42 (1994) 702.
[7] L. Meijer, A. Borgne, O. Mulner, J.P.J. Chong, J.J. Blow, N. Inagaki, M. Inagaki, J.G.
Delcros, J.P. Moulinoux, Eur. J. Biochem. 243 (1997) 527.
[8] K. Torigoe, K. Esumi, Langmuir 9 (1993) 1664.
[9] M. Legraverend, O. Ludwig, E. Bisagni, S. Leclerc, L. Meijer, N. Giocanti, R. Sadri,
V. Favaudon, Bioorg. Med. Chem. 7 (1999) 1281.
[10] C.H. Oh, S.C. Lee, K.S. Lee, E.R. Woo, C.Y. Hong, B.S. Yang, D.J. Baek, J.H. Cho,
Arch. Pharm. Pharm. Med. Chem. 332 (1999) 187.
[11] J.W. Daly, B.E. Christensen, J. Org. Chem. 21 (1956) 177.
[12] P. Imbach, H.G. Capraro, P. Furet, H. Mett, T. Meyer, J. Zimmermann, Bioorg.
Med. Chem. Lett. 9 (1999) 91.
[13] SPARTAN06 (Version 1.1.2), Wavefunction Inc., 18401 Von Karman Avenue, Suite
370, Irvine, CA 92612, USA.
[14] K. Brandenburg, DIAMOND, Release 3.1f, Crystal Impact GbR, Bonn, Germany,
2006.
ˇ
ˇ
ˇ
[15] J. Ulrichová, Z. Dvorák, J. Vicar, J. Lata, J. Smrzová, A. Šedo, V. Šimánek, Toxicol.
Lett. 125 (2001) 125.

1478
P. Štarha et al. / Inorganica Chimica Acta 363 (2010) 1469–1478
[16] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81.
[17] C.J. Pouchert, The Aldrich Library of Infrared Spectra, 3rd ed., Aldrich Chemical
Co., Milwaukee, 1981.
[18] T.A. Mohamed, I.A. Shabaan, W.M. Zoghaib, J. Husband, R.S. Farag, A.E.M.A.
Alajhaz, J. Mol. Struct. 938 (2009) 263.
[19] V. Montoya, J. Pons, J. García-Antón, X. Solans, M. Font-Bardia, J. Ros, Inorg.
Chim. Acta 360 (2007) 625.
[20] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, 5th ed., Willey-Interscience, New York, 1997.
[21] M. Espinal, J. Pons, J. García-Antón, X. Solans, M. Font-Bardia, J. Ros, Inorg.
Chim. Acta 361 (2008) 2648.
[22] Z. Dhaouadi, M. Ghomi, J.C. Austin, R.B. Girling, R.E. Hester, P. Mojzes, L.
Chinsky, P.Y. Turpin, C. Coulombeau, H. Jobic, J. Tomkinson, J. Phys. Chem. 97
(1993) 1074.