Anticancer activity of natural cytokinins: A structure-activity relationship study

Article abstract of DOI:10.1016/j.phytochem.2010.04.018

Cytokinin ribosides (N⁶-substituted adenosine derivatives) have been shown to have anticancer activity both in vitro and in vivo. This study presents the first systematic analysis of the relationship between the chemical structure of cytokinins

Full text of DOI:10.1016/j.phytochem.2010.04.018

Phytochemistry 71 (2010) 1350–1359
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Anticancer activity of natural cytokinins: A structure–activity relationship study
a
a
a
a
a
a
ˇ
ˇ
Jirí Voller , Marek Zatloukal , René Lenobel , Karel Dolezal , Tibor Béreš , Vladimír Kryštof ,
a
b
c
c
a,
*
ˇ
Lukáš Spíchal , Percy Niemann , Petr Dzubák , Marián Hajdúch , Miroslav Strnad
a
´
˚
Laboratory of Growth Regulators, Palacky University and Institute of Experimental Botany ASCR, Šlechtitelu 11, 783 71 Olomouc, Czech Republic
b BIOLOG Life Science Institute, Flughafendamm 9a, 28199 Bremen, Germany
c Laboratory of Experimental Medicine, Palacky´ University, Puškinova 6, 775 20 Olomouc, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Cytokinin ribosides (N6-substituted adenosine derivatives) have been shown to have anticancer activity
both in vitro and in vivo. This study presents the first systematic analysis of the relationship between the
chemical structure of cytokinins and their cytotoxic effects against a panel of human cancer cell lines
with diverse histopathological origins. The results confirm the cytotoxic activity of N6-isopentenyladeno-
sine, kinetin riboside, and N6-benzyladenosine and show that the spectrum of cell lines that are sensitive
to these compounds and their tissues of origin are wider than previously reported. The first evidence that
the hydroxylated aromatic cytokinins (ortho-, meta-, para-topolin riboside) and the isoprenoid cytokinin
cis-zeatin riboside have cytotoxic activities is presented.
Article history:
Received 12 January 2010
Received in revised form 15 April 2010
Accepted 19 April 2010
Available online 1 June 2010
Keywords:
SAR
Cytokinins
Cancer
Most cell lines in the panel showed greatest sensitivity to ortho-topolin riboside (IC50 = 0.5–11.6 lM).
Cytokinin nucleotides, some synthesized for the first time in this study, were usually active in a similar
concentration range to the corresponding ribosides. However, cytokinin free bases, 2-methylthio deriva-
tives and both O- and N-glucosides showed little or no toxicity. Overall the study shows that structural
requirements for cytotoxic activity of cytokinins against human cancer cell lines differ from the require-
ments for their activity in plant bioassays. The potent anticancer activity of ortho-topolin riboside
ortho-Topolin riboside
NCI60
(GI50 = 0.07–84.60 lM, 1st quartile = 0.33 lM, median = 0.65 lM, 3rd quartile = 1.94 lM) was confirmed
using NCI60, a standard panel of 59 cell lines, originating from nine different tissues. Further, the activity
pattern of oTR was distinctly different from those of standard anticancer drugs, suggesting that it has a
unique mechanism of activity. In comparison with standard drugs, oTR showed exceptional cytotoxic
activity against NCI60 cell lines with a mutated p53 tumour suppressor gene. oTR also exhibited signifi-
cant anticancer activity against several tumour models in in vivo hollow fibre assays.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
have only been identified, as yet, in a limited group of plant taxa
(Horgan et al., 1975; Strnad, 1997; Strnad et al., 1992, 1997). The
Cytokinins are important plant hormones that are defined by
their ability to promote cell division in plant tissue culture (Skoog
et al., 1965). Cytokinins found in plants are adenine derivatives
substituted at the N6-position with either isoprenoid or an aro-
matic side chain (Table 1). Isoprenoid trans-zeatin (tZ) is the most
abundant naturally occurring cytokinin. The abundance of other
isoprenoid cytokinins (N6-isopentenyladenine, iP, cis-zeatin, cZ)
and derivatives with a saturated side chain, such as dihydrozeatin
(DHZ), varies between plant species. While isoprenoid cytokinins
are ubiquitous in plants, aromatic cytokinins (represented by N6-
benzyladenine, BA, and its hydroxylated derivatives, the topolins)
most abundant appears to be ortho-topolin riboside, which is
present at micro-molar concentrations in poplar leaves after
daybreak (Hewett and Wareing, 1973). Another aromatic cytoki-
nin, N6-furfuryladenine (kinetin, K), first recognized as a synthetic
compound, has been reported to occur naturally (for a review see
Barciszewski et al. (2007)). Both families of cytokinin occur in
several forms: free bases, ribosides, riboside-50-phosphates,
3-, 7-, 9- and O-glucosides, and amino acid conjugates (Table 1).
The isoprenoid cytokinin iP is an atypical base, present in the tRNA
of all studied organisms, which plays a role in the precise control of
protein synthesis. In mammals, iP is part of tRNA[Ser]Sec and the
cognate tRNA-isopentenyltransferase is a putative tumour sup-
pressor (Spinola et al., 2005). Molecules with a N6-isopentenylade-
nine moiety are released into the cytosol and subsequently into the
body fluids as a result of tRNA turnover (Chheda and Mittelman,
1972).
Abbreviation: PBS, phosphate buffer saline; DTP, developmental therapeutics
program of National Cancer Institute (DTP, Bethesda, USA); SAR, structure–activity
relationship.
* Corresponding author. Tel.: +420 585634850; fax: +420 585634870.
E-mail address: strnad@prfholnt.upol.cz (M. Strnad).
0031-9422/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2010.04.018

J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
1351
Table 1
Structures of the cytokinins used in this study.
R1 NH
N
R5
N
N
R2
N
R4
R3
.
R1
R2
R3
R4
H
Ribosyl
Ribosyl
–
Glucosyl
Ribotide
Ribotide
Ribotide
R5
Trivial name
Abbreviation
H
H
CH3S
H
H
H
H
H
–
–
–
–
–
–
–
–
–
–
–
N6-isopentenyladenine
N6-isopentenyladenosine
2-Methylthio-iPR
iP-7-glucoside
iP-9-glucoside
iPR-50-monophosphate
iPR-50-diphosphate
iPR-50-triphosphate
iP
iPR
2MeSiPR
iP7G
iP9G
iPRMP
iPRDP
iPRTP
CH3
CH3
CH2
Glucosyl
–
–
–
–
H
H
H
H
H
CH3S
–
–
–
–
–
–
H
–
–
trans-Zeatin
t-Zeatin riboside
t-Zeatin-7-glucoside
t-Zeatin-9-glucoside
tZR-50-monophosphate
2-Methylthio-t-zeatin riboside
tZ
tZR
tZ7G
tZ9G
tZRMP
2MeStZR
CH2OH
CH3
Ribosyl
–
Glucosyl
Ribotide
Ribosyl
Glucosyl
CH2
–
–
–
H
H
–
–
H
–
–
cis-Zeatin
cis-Zeatin riboside
cZ
cZR
CH3
Ribosyl
CH2
CH2
CH2
CH2
CH2OH
CH2OG
CH3
H
H
–
–
H
–
–
t-Zeatin-O-glucoside
t-Zeatin riboside-O-glucoside
tZOG
tZROG
Ribosyl
H
–
H
–
t-Zeatin-O-acetyl
ActZ
CH2OAc
CH3
H
H
H
H
H
–
–
–
–
–
H
–
–
Dihydrozeatin
DHZ
DHZR
DHZ7G
DHZ9G
DHZRMP
CH2OH
CH3
Ribosyl
–
Glucosyl
Ribotide
Dihydrozeatin riboside
Dihydrozeatin-7-glucoside
Dihydrozeatin-9-glucoside
DHZR-50-monophosphate
Glucosyl
–
–
H
H
–
–
H
–
–
Dihydrozeatin-O-glucoside
Dihydrozeatin riboside-O-glucoside
DHZOG
DHZROG
CH2OG
CH3
Ribosyl
CH2
–
–
–
–
–
–
–
–
–
–
H
–
–
–
N6-benzyladenine
BA
BAR
N6-benzyladenosine
CH2
Ribosyl
–
–
Glucosyl
Ribotide
Ribotide
Ribotide
Glucosyl
N6-benzyladenine-3-glucoside
N6-benzyladenine-7-glucoside
N6-benzyladenine-9-glucoside
BAR-50-monophosphate
BAR-50-diphosphate
BA3G
BA7G
BA9G
BARMP
BARDP
BARTP
–
–
–
–
–
Glucosyl
–
–
–
–
BAR-50-triphosphate
–
–
–
–
–
–
H
–
–
–
meta-Topolin
meta-Topolin riboside
meta-Topolin-9-glucoside
mT
mTR
mT9G
CH2
Ribosyl
Glucosyl
HO
–
–
–
–
–
–
CH3S
–
–
–
–
–
–
–
H
–
–
–
–
–
–
–
ortho-Topolin
oT
oTR
oT9G
oTRMP
oTRDP
oTRTP
2MeSoTR
CH2
Ribosyl
Glucosyl
Ribotide
Ribotide
Ribotide
Ribosyl
ortho-Topolin riboside
ortho-Topolin-9-glucoside
oTR -50-monophosphate
oTR -50-diphosphate
oTR -50-triphosphate
2-Methylthio-oTR
OH
H
H
H
–
–
–
H
Kinetin
Kinetin riboside
KR -50-monophosphate
K
KR
KRMP
O
CH2
Ribosyl
Ribotide
Knowledge that cytokinins play key roles in the regulation of
ability of cytokinin bases to induce or promote the differentiation
of human cells has been demonstrated in both keratinocytes (Berge
et al., 2006) and several leukaemia cell lines, including HL-60 and
K-562 (Ishii et al., 2003).
However, while cytokinin bases induce differentiation at rela-
tively high concentrations (25–100 lM), their ribosides cause rapid
plant growth and differentiation led to postulation that they could
also affect growth and differentiation in animals, and have poten-
tial utility for treating human diseases that involve dysfunctional
cell proliferation and/or differentiation. Abundant evidence sup-
porting these hypotheses has subsequently been obtained. The

1352
J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
apoptosis of leukaemia cell lines at low micro-molar concentra-
tions (Mlejnek, 2001). Cell death in HL-60 is preceded by depletion
of adenosine triphosphate, activation of caspases and mitochon-
drial depolarization (Mlejnek, 2001; Ishii et al., 2002). Intracellular
conversion of ribosides to their respective monophosphates is
However, only a few cytokinins have been tested as yet for antican-
cer activity. Here, we report an analysis of the relationship be-
tween the structures of 47 cytokinins (Table 1) and their
cytotoxic activity against a panel of cell lines derived from diverse
malignancies (Table 2). The tested compounds include almost all
known natural purine cytokinins, representing all structural
variants. Several compounds were synthesized for the first time
for this study, including 2-methylthio-ortho-topolin riboside
(2MeSoTR), ortho-topolin riboside-5’-monophosphate (oTRMP)
and selected cytokinin riboside di- and triphosphates. Cytotoxic
activity was evaluated after 72-h treatments using a standard
viability test, based on quantification of the fluorescent product
of the enzymatic hydrolysis of Calcein AM. The determined activi-
ties were expressed as IC50 values (concentrations leading to a 50%
decrease in cellular esterase activity).
ˇ
known to be required for their action (Mlejnek and Dolezel,
2005). It has recently been demonstrated that kinetin riboside
(KR) is a potential drug for the treatment of multiple myelomas
(Tiedemann et al., 2008). In several models of multiple myeloma,
KR has been found to induce rapid suppression of cyclin D1 and
D2 transcription, followed by arrest of the cell-cycle and selective
apoptosis in tumour cells (Tiedemann et al., 2008). Several authors
have reported cytotoxic effects of N6-isopentenyladenosine (iPR),
KR and N6-benzyladenosine (BAR) on human cell lines derived
from solid tumours (Cabello et al., 2009; Choi et al., 2008; Griffaut
et al., 2004; Laezza et al., 2009; Meisel et al., 1998; Spinola et al.,
2007). Whether treatments resulted in cell cycle block and/or
apoptosis was dependent on the cell line and the cytokinin used.
The anticancer activity of iPR, KR and BAR has been demonstrated
in vivo using several animal and xenograft models of cancer (Choi
et al., 2008; Laezza et al., 2006; Tiedemann et al., 2008). iPR and
BAR have also shown promising activity against a diverse range
of cancers in a limited clinical trial (Mittelman et al., 1975).
Micro-molar concentrations of both cytokinin ribosides and
cytokinin bases can also induce cell death in plant cell cultures,
with some identifiable characteristics of apoptosis (activation of
caspase-like proteases and fragmentation of DNA) (Mlejnek and
Procházka, 2002). This cell death is preceded by depletion of aden-
osine triphosphate and the production of reactive oxygen species.
In contrast to their hormonal activity, which requires interaction
with specific membrane-bound receptors, intracellular conversion
of cytokinins to monophosphates is necessary for this cytotoxic ef-
fect. The concentrations of cytokinin required to produce cytotoxic
effects are higher than those found endogenously in plant tissues,
but they do fall within the range used in plant bioassays (Carimi
et al., 2003; Mlejnek et al., 2003, 2005).
Treatment with the ribosides, oTR, iPR, BAR and KR resulted in a
dose-dependent reduction in the viability (IC50 = 0.5–13.6 lM) of
the cell lines derived from both haematological malignancies
(CEM, HL60, K562 and RPMI) and solid tumours (MCF7, HeLa,
and HOS). The leukaemia cell lines, HL60 and CEM, were most sen-
sitive, with determined IC50 values 61.7
across the whole panel of cancer cell lines were 2.27, 4.15, 5.10
and 5.95 M for oTR, BAR, KR and iPR, respectively. While the cyto-
lM. Median IC50 values
l
toxic activity of the cytokinin ribosides iPR, BAR and KR against
cancer cell lines has already been reported (Ishii et al., 2002;
ˇ
Laezza et al., 2009; Meisel et al., 1998; Mlejnek and Dolezel,
2005; Spinola et al., 2007) this is the first report of the activity of
oTR. The findings that cytokinin ribosides are active at sub-mi-
cro-molar (some leukemias) or low micro-molar concentrations
(other leukemias, adherent cells) are consistent with those of pre-
vious studies, despite variations in the design of the cytotoxicity
assays used in terms of principle/mechanism, definition of the end-
point and length of treatment. In contrast to oTR, the meta- and
para-isomers only exhibited inhibitory activity against HL60 (with
IC50 values of 24 and 7.5 lM, respectively). Among the cytokinin
ribosides with hydroxylated isoprenoid side chains, cis-zeatin ribo-
side (cZR) but not tZR or dihydrozeatin riboside (DHZR), was active
against the leukaemia cell lines CEM and HL60 (with IC50 values of
Although the cytotoxic activity of natural cytokinins and their
ˇ
analogues (Dolezal et al., 2006, 2007) against mammalian cells
has been repeatedly demonstrated, there have been no previously
published systematic studies of the structure-cytotoxic activity
relationship (SAR) of natural cytokinins. Available reports indicate
that only KR, iPR, BAR, trans-zeatin riboside (tZR) and their free
bases have been tested, to date. Therefore, the present study was
undertaken to examine the cytotoxic effects of almost all known
naturally occurring cytokinins against a panel of cancer cell lines
of diverse histopathological origin and then determine the basic
SAR in terms of their growth-inhibitory effects. Furthermore, the
activity pattern of ortho-topolin riboside (oTR) against NCI60, a
thoroughly characterized panel of 59 human cancer cell lines
(Shoemaker, 2006) was analysed. Finally, we report results of
in vivo tests of the anticancer activity of oTR against models repre-
sentative of human tumours in hollow fibre assays.
18.8 and 7.9 lM, respectively).
Compared to the tested cytokinin ribosides, the free bases typ-
ically had much weaker effects on cell proliferation, with IC50 val-
ues either over the highest concentration used (>166 lM) or at
least 50 times higher than the IC50 values determined for their
respective ribosides. Similar differences between the cytotoxic
activity of cytokinin bases (K, iP, and BA) and their corresponding
ribosides have been reported by other authors (Ishii et al., 2002;
ˇ
Mlejnek and Dolezel, 2005). In cultures of plant cells, the toxicities
of cytokinin bases and the corresponding ribosides are comparable,
because in contrast to human cells plant cells can convert both
forms of cytokinin efficiently into riboside-50-monophosphates
(Mlejnek et al., 2003, 2005). Therefore, it was postulated that low
toxicity of K, iP and BAP in human leukaemia HL-60 cell line is
due to the low activity of human phosphoribosyltransferase to-
ˇ
wards cytokinin bases (Mlejnek and Dolezel, 2005). The differences
2. Results and discussion
observed in the cytotoxic activities of ortho-topolin (oT), para-top-
olin (pT), meta-topolin (mT) and cis-zeatin (cZ) compared to their
corresponding ribosides may have similar causes.
With the exception of trans-zeatin riboside-O-glucoside
(tZROG), which exhibited some activity against the leukaemia cell
In addition to their essential roles in the growth and develop-
ment of plants, cytokinins have various effects in man and animals
at both cellular and whole organism levels (Berge et al., 2006; Sla-
ugenhaupt et al., 2004; Rattan and Sodagam, 2005; Wu et al.,
2007). Hence, cytokinins and their derivatives have many potential
therapeutic applications, including possible efficacy in the treat-
ment of proliferative diseases such as cancers. The anticancerous
activity of cytokinins in a variety of cultured cell lines, several
xenografts and even a clinical trial has been documented (Ishii
et al., 2002; Mittelman et al., 1975; Tiedemann et al., 2008).
lines CEM and HL-60 (IC50 ꢀ 26
l
M), cytokinin O- and N-glucosides
M). The observation that
were inactive in the assay (IC50 > 166
l
cytokinin bases and cytokinin glucosides showed limited activity,
or none at all, supports the hypothesis that the presence of a ribose
moiety at N9 of the purine ring is essential for potent anticancer
activity in cytokinins. A decrease in the cytotoxic activity of two or-
ders of magnitude was also observed after substitution of oTR and

J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
1353
Table 2
Antiproliferative activity of cytokinins expressed as IC50 values in a 3-day Calcein-AM assay. Presented values are averages of at least three independent experiments, where
individual replicate values fell within 25% of the average.
CEM
HL60
K562
140
RPMI 8226
>166
HOS
MCF7
>166
G361
>166
HELA
>166
BJ
BA
>166
>166
>166
>166
1.3
1.1
2.0
2.3
>166
>166
>166
>166
0.93
0.95
1.3
>166
>166
>166
>166
13.6
>166
BA7G
BA9G
BA3G
BAR
5.9
3.7
4.0
3.9
4.6
5.3
4.5
4.7
3.7
5.0
8.0
11
15.0
>166
1.9
2.3
1.7
2.6
BARMP
BARDP
BARTP
K
K9G
KR
KRMP
iP
iP9G
14.4
4
155
>166
1.6
1.8
92
>166
1.7
2.4
2.5
3.4
119
118
>166
0.5
1.3
>166
>166
0.8
>166
>166
>166
>166
5.9
11.9
>166
>166
11.7
5.1
>166
>166
>166
>166
2.1
10.5
13.6
150
7.4
16.5
>166
4.3
4.3
>166
21.9
38.8
>166
4.3
12.2
>166
1
>166
>166
0.71
1.2
1.4
2.8
85.5
78
>166
0.6
0.48
1.9
1.6
>166
>166
24
>166
123
7.5
>166
>166
>166
>166
>166
26.3
95.9
78.6
>166
>166
>166
7.9
26.4
>166
>166
>166
>166
>166
iPR
5.2
4.3
5.3
6.6
>166
>166
6.4
5.3
5.4
7
>166
>166
6.9
2.8
3.5
5.2
>166
>166
>166
>166
5.5
4.2
2.5
2.1
iPRMP
iPRDP
iPRTP
2MeSiPR
oT
>166
>166
>166
3.1
2.5
1.9
>166
103
>166
120
>166
>166
oT9G
oTR
2.4
4.0
2.1
3.5
>166
140
>166
1.5
3.0
1.9
4.3
130
>166
>166
7.7
11.6
13.5
11.9
17.2
155
2.13
4.5
1.5
1.2
oTRMP
oTRDP
oTRTP
2MeSoTR
mT
mTR
mT9G
pT
pTR
pT9G
tZ
tZ7G
tZ9G
tZOG
tZROG
tZR
10.5
12.1
14.4
>166
>166
>166
0.8
1.6
2.7
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
27.5
>166
>166
>166
>166
>166
18.8
61
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
3.9
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
15.3
>166
>166
>166
>166
>166
120
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
tZRMP
ActZ
2MeStZR
cZ
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
>166
42.3
>166
>166
>166
>166
>166
>166
55.2
>166
>166
>166
cZR
cZRMP
2MeScZR
DHZR
DHZROG
DHZMP
>166
>166
151
>166
>166
iPR at C2 with a 2-methylthio group. This, along with other substi-
tutions that also led to attenuation of the cytotoxic effects of cyto-
kinin ribosides could be useful in programmes to develop drugs to
treat conditions other than cancers. It will be of interest to estab-
lish whether these observed decreases in cytotoxic activity are
the result of reduced affinity of adenosine kinase for C2-substi-
tuted cytokinin ribosides.
The present SAR study illustrates that the structural require-
ments for cytokinins to show potent cytotoxic activity against hu-
man cancer cell lines are different from those needed for their
activity in plant bioassays. While the ribose moiety appears to be
important for cytotoxic effects against a diverse range of cancer
cells, conjugation of cytokinin bases with sugars, including ribose,
serves to limit cytokinin signalling in plant cells. The hydroxyl
position of the side chain of cytokinins had a marked effect on anti-
cancer activity, in both aromatic (oTR >> meta-topolin riboside –
mTR, para-topolin riboside – pTR) and isoprenoid (cZR >> tZR)
cytokinins. In contrast, tZ and mT cytokinins are more active than
their positional isomers (cZR, oTR, pTR) in plant bioassays (Holub
et al., 1998; Letham and Palni, 1983; Spíchal et al., 2004).
Extracellular adenosine nucleotides play important roles in a di-
verse range of physiological processes, including cell death in
mammals (reviewed by Burnstock (2007), for example). In addi-
tion, information is available on: the release of intracellular ade-
nine nucleotides; their inter-conversion (cycling) in the blood
and intercellular space by both membrane-bound and soluble en-
zymes; and specific signalling pathways (Yegutkin, 2008; Yegutkin
et al., 2003). Given their importance we synthesized 50-nucleotides
of selected cytokinin ribosides and examined their activity in var-
ious tests, including the Calcein-AM viability assay.
Phosphorylation of purine ribosides also increases the solubility
of nucleosides, which can lead to improvements in both drug for-
mulation and bioavailability. The practicality of this approach has
been demonstrated by the development of fludarabine (9-b-D-ara-
binosyl-2-fluoroadenine-5’-monophosphate), the active ingredient
in the drug FludaraÒ, which has been approved for the treatment of
certain haematological malignancies (Anderson and Perry, 2007).
Cytokinin riboside-50-phosphates derived from oTR, iPR, KR,
BAR and cZR markedly reduced the growth of cell lines that were
sensitive to the parent compounds. In most cases the activity of

1354
J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
Fig. 1. Identification of intracellular oTRMP by capillary electrophoresis with UV detection (268 nm). CEM cells were treated with 25 lM oTRMP for 3 h. The inset window
illustrates separation of the standards.
the ribosides and the respective riboside-50-phosphates was either
similar, or phosphorylation led to decreased activity. Simultaneous
Due to its drug-like properties and the promising biological
activity it demonstrated, oTR was selected for further testing
against the NCI60 cancer cell line panel at the developmental ther-
apeutics program, Division of Cancer Treatment and Diagnosis, Na-
tional Cancer Institute (DTP, Bethesda, USA). NCI60 is a collection of
59 human cancer cell lines that have been extensively character-
ized at the DNA, RNA and protein levels and are used for routine
drug screening at DTP. Comparisons of patterns of activity (GI50
values of a compound for individual NCI60 cell lines) by, for exam-
ple, Pearson correlation, may be useful for identifying compounds
that could have common mechanisms of action. Similarly, compar-
ison of activity patterns and expression patterns of molecular tar-
gets provides a means to detect molecular markers that influence
the cells’ sensitivity to a compound (Shoemaker, 2006).
treatment with an adenosine kinase inhibitor, A-134974 (5
lM),
protected CEM and K562 cells (IC50 > 100 M) not only against
l
oTR and iPR, but also against their respective mono-, di- and tri-
phosphates. Therefore, intracellular phoshorylation appears to be
an important step in the sequence of events leading to cytotoxicity
after the application of not only cytokinin ribosides (Cabello et al.,
ˇ
2009; Mlejnek and Dolezel, 2005) but also their respective ribo-
side-50-phosphates. These observations suggest that, similarly
to other nucleotide drugs, such as fludarabine monophosphate
(Malspeis et al., 1990) and triciribine monophosphate (Wotring
et al., 1986), cytokinin riboside-50-phosphates are dephosphoryl-
ated extracellularly and the resulting nucleosides are transported
across the membrane and then re-phosphorylated in the cell.
Intracellular accumulation of monophosphorylated oTR after the
The cytotoxic activity of oTR against the NCI60 panel is shown in
Fig. 2, in terms of GI50 values (concentrations causing a 50% reduc-
tion in cell growth). oTR was potently active against most of the
treatment of CEM cells with oTRMP (25 lM, 3 h) was demon-
strated by capillary electrophoresis (see Fig. 1).
cell lines, with a median GI50 value of 0.65
two lines, GI50 values <10 M. Similarities in the median GI50 val-
ues (range 0.34–1.32 M) together with generally high degrees of
lM and, for all except
The intracellular concentration was determined to be between
0.67 and 0.92 mM (in three biological replicates), exceeding the
concentration applied to cells more than 25-fold. The identity of
oTRMP was confirmed by spiking the sample with a standard
solution of the compound. If adenosine kinase was inhibited by
A-134974 or the experiment was carried out in serum-free med-
ium, concentrations of oTRMP were found to be below the limit
of detection. The peak of oTRMP was missing when the cells were
treated with DMSO vehiculum (data not shown).
In order to obtain preliminary information about the selectivity
of cytokinin nucleosides and nucleotides, their cytotoxic effects
against human BJ fibroblasts (as a model primary cell line) were
also examined. Most of the compounds tested, including those pre-
viously used in mouse xenograft experiments (iPR, BAR, KR) and
clinical trials (iPR, BAR) showed significant cytotoxicity, with IC50
values frequently in the low micro-molar range (Table 2). Future
in vivo experiments will be required to demonstrate whether there
is a therapeutic window for cytotoxic cytokinins. The results of the
mouse hollow fibre assay described below, demonstrating the
in vivo anticancer activity of oTR against implanted human tu-
mours at concentrations causing no acute toxicity, indicate that
this is a possibility.
l
l
variability within tissue sub-groups (max/min ratio >15 in all
sub-panels except the ovarian and prostate sub-groups) suggest
that factors other than the tissue origin determine sensitivity of
cancer cells to oTR. Below we analyse the effects of p53 status of
the cells and adenosine kinase expression on the activity of oTR
and compare its activity pattern with those of 214 antineoplastics
(‘‘clinical agents”) that have already been clinically evaluated.
Analysis of the influence of the mutational status of the NCI60
cell lines (Ikediobi et al., 2006) on their sensitivity to oTR led to
the important observation that cell lines carrying the mutant p53
tumour suppressor gene (median GI50 = 0.60
more sensitive than cells with the wild type variant (median
GI50 = 1.59 M). TP53, the protein product of this gene, is a vital
lM) are generally
l
component of the regulatory system that responds to various cell
stressors, including DNA damage, oncogene activation, hypoxia,
and disruption of cell adhesion. Activation of TP53 can result in cell
cycle arrest, senescence and apoptosis. The p53 gene is known to
be either mutated or deleted in over 50% of all human cancers
(Vazquez et al., 2008). Dysfunction of TP53 has also been impli-
cated in chemo- and radio-resistance (Bossi and Sacchi, 2007;

J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
1355
Fig. 3. Histogram showing the relationship between p53 status and activity of oTR
(black bar) and 214 clinically evaluated compounds. The strength of each
relationship is expressed as a p-value derived from the Wilcoxon rank sum test.
The alternative hypothesis was that the individual compounds are more active in
cell lines with mutated p53.
oTR. Further studies with an independent cancer cell panel are re-
quired to test this hypothesis.
Fig. 2. Negative log GI50 values (M) for individual tissue types in NCI60. Grey lines
indicate the global and tissue-specific median. BR – breast cancer, CNS – central
nervous system cancer, CO – colon cancer, LC – non-small cell lung cancer, LE –
leukaemia, ME – melanoma, OV – ovarian cancer, PR – prostate cancer and RE –
renal cancer.
In order to identify compounds that have a similar pattern of
activity (and possibly, therefore, a similar mechanism of action)
we calculated Pearson correlation coefficients for patterns of activ-
ity of oTR and ‘‘clinical agents”. Compounds were ranked using
respective p-values. Only an inhibitor of ribonucleotidase, carace-
mide (Moore and Loo, 1984) had a correlation coefficient that
was higher than 0.4 (r = 0.42, p = 0.0013). Further positive correla-
tions (r > 0.35, p < 0.005) were obtained for the purine anti-metab-
olites (3-deazaguanine and diglycoaldehyde), the cAMP analogue
8Cl-cAMP, and the tricyclic ribotide Akt inhibitor triciribine mono-
phosphate (Cory et al., 1976; Kim et al., 2005; Pieper et al., 1988).
None of these compounds is currently approved for use in humans.
Although statistical significances (defined as p < 0.05) of these cor-
relations did not survive Bonferroni correction for multiple testing,
the relations reported here may have biological meaning. Correla-
tions between the patterns of activity of oTR and the purine ana-
logues may reflect the ability of oTR to interact with the human
purinome and, more specifically, with enzymes involved in purine
and nucleic acid metabolism. Triciribine monophosphate, a pro-
drug of triciribine, is known to be dephosphorylated extracellu-
larly, transported into cells and then re-phosphorylated by
adenosine kinase (Wotring et al., 1986). Here, we propose a similar
mechanism of internalisation and metabolic activation for the
cytokinin nucleotides. The correlation between the GI50 patterns
of oTR and triciribine monophosphate might, therefore, reflect
the importance of adenosine kinase in the metabolism of both
drugs. We conclude that since the proportion of variance shared
by the activity patterns of oTR and each individual ‘‘clinical agent”
was always lower than 18% (r2 < 0.18), a unique combination of
biological factors is probably underlying the sensitivity of the
NCI60 cell lines to oTR.
Weller, 1998). The potential significance of the strong activity of
oTR towards mutant p53 lines is highlighted by the report that
most of nearly 90 clinically evaluated anticancer agents showed
greater activity against NCI60 cell lines carrying wild-type p53
(Weinstein et al., 1997). We used the same approach as that ap-
plied by Weinstein et al. to compare the effect of p53 status on
the growth inhibitory activity (GI50) of oTR and 214 standard
antineoplastics. This set of ‘‘clinical agents” was created by pooling
antineoplastics from the ‘‘Approved Oncology Drugs” and ‘‘Stan-
dard Agents” DTP datasets (see Section 4 for details). p-Values ob-
tained from one-sided Wilcoxon rank sum tests comparing the GI50
values of individual compounds on the cell lines with mutated and
wild-type p53 were used as a metrics of the dependence of the
activity on p53 status. The resulting distribution of p-values is
shown in Fig. 3. While most of the compounds tested showed
more activity against the cells with wild-type p53 (indicated by
p-values > 0.5) oTR (p-value = 0.035, rank 2) was exceptionally
active against the cells with the mutant p53 gene.
Previous studies on cytokinin ribosides have demonstrated that
intracellular phosphorylation by adenosine kinase is a requirement
ˇ
for a cytotoxic effect (Mlejnek and Dolezel, 2005) and, as men-
tioned above, the cytotoxic effect of oTR can be prevented by treat-
ment with an adenosine kinase inhibitor. Therefore, it was of
interest to determine whether a relationship between the activity
of oTR and expression of adenosine kinase exists. The GI50 values
of oTR were found to be negatively correlated (r < ꢁ0.43,
p < 0.0006) with signals of both adenosine kinase probes on
U133A Affymetrix expression arrays in the Genelogic dataset
(Shankavaram et al., 2007). Significant negative correlations be-
tween the GI50 values of oTR and adenosine kinase expression
(r < ꢁ0.69, p < 0.019) were also observed in the melanoma sub-
group. Therefore, it is possible that adenosine kinase expression
in cells could be used as a biomarker of the sensitivity and/or
resistance of certain malignancies (for example, melanoma) to
Finally, the activity of oTR against tumours derived from 12
NCI60 cell lines was tested in vivo in hollow fibre assays. oTR was
administered by intraperitoneal injection for four consecutive days
at two dose levels, 100 and 150 mg/kg/day, which were found to be
safe in a preliminary acute toxicity study (data not shown). oTR
caused a 50% or greater reduction (as measured by a standard
MTT assay) of tumour mass in 16 out of 24 intraperitoneal
implants, resulting in an ip score of 32 out of 48. No tumour

1356
J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
reduction exceeding threshold assay values was achieved when the
drug was injected subcutaneously (sc score 0). Ip score (but not sc
score) was shown to be positively correlated with the likelihood of
activity in xenograft models. Notably, the relation was stronger
when both intra- and extraperitoneal grafts were considered
(Johnson et al., 2001; Decker et al., 2004). According to DTP meth-
odology, compounds with an ip score greater than 7 or total
(ip + sc) score greater than 19 are considered as candidates for fol-
low-up xenograft experiments.
BARMP, K9G, KRMP, iP9G, iPRMP, 2MeSiPR, cZ, tZ7G, tZ9G, tZRMP,
tZOG, tZROG, 2MeStZR, cZR, cZRMP, 2MeScZR, ( )DZR, DZROG and
DZMP were generous gifts from Olchemim Ltd. (Olomouc, Czech
Republic). The aromatic cytokinins oT, oTR, oT9G, mT, mTR,
mT9G, pT, pTR and pT9G were synthesized, according to proce-
dures described by Holub et al. (1998). Before they were used
the purity of all cytokinins was tested by HPLC (Strnad et al.,
1997). Dulbecco’s modified Eagle’s medium (DMEM), RPMI 1640,
fetal bovine serum (FBS), L-glutamine, penicillin, streptomycin
While the high ip score shows promising anticancer activity of
oTR at the site of application, the absence of the effect on the sub-
cutaneous implants points to the limited systemic availability of
the drug. Other routes of application might be more appropriate
for the relatively polar cytokinin ribosides. In this context, it would
be interesting to compare the ip and sc scores of other nucleoside
analogs. In humans, significant differences in peritoneal and plas-
ma exposure were observed after intraperitoneal application of
cytarabine and gemcitabine. Both the drugs were successfully used
in high dose regional therapy of intra-abdominal cancers (King
et al., 1984; Kamath et al., 2009). Follow-up mouse xenograft
experiments (testing various sites of implantation and modes of
application) are necessary to further evaluate potential utility of
oTR in cancer therapy.
and A-134974 were purchased from Sigma (MO, USA) and Calcein
AM from PAA Laboratories GmbH (Pasching, Austria). All reagents
used were either of analytical grade or the highest grade available
from commercial suppliers.
4.2. Cytokinin nucleotide synthesis
All N6-substituted adenosine-50-O-di- and tri-phosphates used
in the present study were synthesized by treating the triethylam-
monium salt of the corresponding 6-chloropurine-9b-D-riboside-
50-O-di- or triphosphate precursors with appropriate side chain
amines under aqueous, alkaline conditions. The reactions were
monitored by analytical chromatography using 250 ꢂ 4.6 mm
stainless steel column packed with Gemini™ C18–110, 5 lm parti-
cles (Phenomenex, Aschaffenburg, Germany) as solid phase. The
mobile phase was 25% CH3CN, 25 mM Na2HPO4, 4 mM tributylam-
monium sulphate (pH 7). The flow rate was 1.25 ml/min. The
products were partially purified by preparative MPLC with DEAE-
3. Concluding remarks
The ability of the plant hormones cytokinins to induce apopto-
sis and/or block cell cycling in a wide range of cancer cells makes
them potential candidates as drugs for treating a variety of cancers.
This study represents the first systematic analysis of the relation-
ship between the structure of cytokinins and their cytotoxic ef-
fects, assessed using a diverse panel of human cancer cell lines.
The results confirm the anticancer activity of iPR, KR and BAR,
and demonstrate that the range of cell lines sensitive to cytokinins
and the tissue origin of these cells is wider than previously re-
ported. In addition, the anticancer activity of the hydroxylated aro-
matic (oTR, pTR, mTR) and isoprenoid cytokinins (cZR) is reported
for the first time. Against most cell lines tested, ortho-topolin ribo-
Sepharose (125 ꢂ 25 mm glass column of Q Sepharose FF, 90
lm,
Amersham Biosciences, Freiburg, Germany) as the stationary phase
and 300 mM triethylammonium bicarbonate (pH 8) as the mobile
phase. The flow rate was 10 ml/min. The fractions containing the
required product were collected, concentrated under reduced
pressure, and then desalted using preparative HPLC equipped with
220 ꢂ 50 mm stainless steel column packed with ODS-AQ™
C18–100, 16 lm (YMC Europe/Sinsheim/Germany) at the flow rate
of 10 ml/min. The purified compounds were eluted with 5% MeOH,
concentrated at reduced pressure, and stored as 10 mM aqueous
stock solutions at ꢁ70 °C.
The analytical HPLC system for the monitoring of product for-
mation progress and its purity as well as the preparative HPLC sys-
tem for product purification consisted of a L7100 pump, a L7400
side (IC50 = 0.5–11.6 lM) was the most active cytokinin. Cytokinin
nucleotides (some synthesized for the first time in this study) were
active against the same cell lines as the parent ribosides. Cytokinin
free bases, including 2-methylthio derivatives as well as O- and
N-glucosides, exhibited limited toxicity or none at all. It can be
concluded from this study that cytokinins have different structural
requirements for cytotoxic activity against human cancer cell lines
than for activity in plant bioassays. The potent anticancer activity
of oTR was confirmed through further testing of this compound
variable wavelength UV-detector, and
a D 7500 Integrator
(Merck-Hitachi, Darmstadt, Germany). Mass spectra of the reaction
products by ESI-MS were measured in isopropanol–H2O–HCOOH
(50:49.9:0.1, v/v/v). Helios b spectrometer (Spectronic Unicam,
Cambridge, UK) was used to record UV-spectra of the compounds
in aqueous phosphate buffer, pH 7.
on NCI60 (median GI50 = 0.65 lM), a standard panel of 59 cell lines
4.2.1. N6-(
D
2-Isopentenyl)adenosine-5’-O-diphosphate (iPRDP)
-riboside-5’-
originating from nine different tissues. The activity pattern deter-
mined for oTR against NCI60 could be clearly distinguished from
the patterns of a set of standard, established anticancer drugs, sug-
gesting that a unique combination of factors underlie its activity.
Another significant finding was the high sensitivity of the NCI60 cell
lines with mutated p53 tumour suppressor gene to oTR and signif-
icant differential toxicity of oTR in tissue origin sub-panels. oTR is
toxic to rapidly dividing normal diploid fibroblasts in vitro, but was
found to have significant anticancer activity in several tumour
models in vivo at concentrations causing no acute toxicity.
iPRDP was synthesized from 6-chloropurine-9b-
D
O-diphosphate, triethylammonium salt, and 2-isopentenylamine
by nucleophilic substitution with 2-isopentenylamine in the pres-
ence of sodium hydroxide (pH 11) in water at 40 °C. The formation
of the product was monitored by analytical HPLC. The reaction was
quenched by adding formic acid and subsequently cooling to
ꢁ70 °C, then the product was purified by preparative MPLC with
DEAE Sepharose as the stationary phase and 300 mM triethylam-
monium bicarbonate (pH 8) as the mobile phase. Fractions con-
taining product were collected, concentrated under reduced
pressure, and desalted by preparative HPLC (trapping on reversed
phase silica gel, washing with water then eluting with 5% MeOH).
The triethylammonium salt of iPRDP was isolated with >98% purity
(yield: 32%). iPRDP triethylammonium salt: white solid; UV–VIS
4. Experimental
4.1. Materials
(phosphate buffer pH 7.0) kmax (log e) 268 (4.228) nm; ESI-MS
The cytokinins iP, iPR, tZ, tZR, DHZ, BA, BAR, K, and KR were pur-
chased from Sigma (St. Louis, MO). ActZ, BA3G, BA7G, BA9G,
(pos.) m/z: 494.1 ([M+H]+); ESI-MS (neg.) m/z: 496.1 ([MꢁH]ꢁ);
Mr calculated for the free acid (C15H23N5O10P2): 495.32.

J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
1357
4.2.2. N6-(
D
2-Isopentenyl)adenosine-5’-O-triphosphate (iPRTP)
-riboside-5’-O-
were collected, concentrated under reduced pressure, and desalted
by preparative HPLC (trapping on reversed phase silica gel, wash-
ing with water then eluting with 5% MeOH). oTRTP was isolated
as the triethylammonium salt with a purity of >97% (yield: 42%).
oTRTP triethylammonium salt: white solid; UV–VIS (phosphate
iPRTP was synthesized from 6-chloropurine-9b-
D
triphosphate, triethylammonium salt, and isopentenylamine by
nucleophilic substitution with 2-isopentenylamine in the presence
of sodium hydroxide (pH 11) in water at 40 °C. Again, product for-
mation was monitored by analytical HPLC. The reaction was
quenched by adding formic acid and subsequent cooling to
ꢁ70 °C, the product was purified by preparative MPLC with DEAE
Sepharose as the stationary phase, using 300 mM triethylammo-
nium bicarbonate (pH 8) as the mobile phase. Again, fractions con-
taining product were collected, concentrated under reduced
pressure, and desalted by preparative HPLC (trapping on reversed
phase silica gel, washing with water then eluting with 5% MeOH).
The triethylammonium salt of iPRTP was isolated with a purity
of >98% (yield: 42%). iPRTP triethylammonium salt: white solid;
buffer pH 7.0) kmax (log e) 270 (4.312) nm; ESI-MS (pos.) m/z:
614.0 ([M+H]+); ESI-MS (neg.) m/z: 612.0 ([MꢁH; Mr calculated
for the free acid (C17H22N5O14P3): 613.35.
4.2.6. 2-Methylthio-6-(2-hydroxybenzyl)aminopurine riboside
(2MeSoTR)
2MeSoTR was prepared from 2-methylthio-6-chloropurine-9b-
D-riboside and 2-hydroxybenzylamine. The reaction was carried
out, at 90 °C for 20 h, in methanol in the presence of triethylamine
in nitrogen gas. The reaction mixture was then evaporated in a vac-
uum evaporator and residue that was insoluble in 25% methanol
was re-crystallized from 70% methanol. Purity of the final product
was determined to be 97% (HPLC) and the calculated yield 50.5%.
2MeSoTR: white solid; ESI-MS (pos.) m/z: 420.3 ([M+H]+); Mr
calculated for the free acid (C18H21N5O5S): 419.46.
UV–VIS (phosphate buffer pH 7.0) kmax (log e) 268 (4.228) nm;
ESI-MS (pos.) m/z: 576.1 ([M+H]+); ESI-MS (neg.) m/z: 574.2
([MꢁH]ꢁ); Mr calculated for the free acid (C15H24N5O13P3): 575.30.
4.2.3. N6-(2-Hydroxybenzyl)aminopurine riboside-5’-O-
monophosphate (oTRMP)
oTRMP was prepared from 6-chloropurine-9b-D-riboside-5’-O-
monophosphate, disodium salt dihydrate, by nucleophilic substitu-
tion with 2-hydroxybenzylamine in the presence of N,N-diisopro-
pyl-N-ethylamine in methanol. The reaction was carried out at
90 °C for 12 h in a nitrogen atmosphere. The solvent was removed
by evaporation under vacuum and raw oTRMP was purified by RP
C18 flash chromatography (mobile phase 15% methanol in water)
followed by crystallization from propan-2-ol. The purity of the
final product was 95% (HPLC) and the yield 80%. oTRMP sodium
4.3. Capillary electrophoresis
CEM cells were harvested by centrifugation (500g, 4 °C, 5 min),
washed twice in an excess of ice cold phosphate buffered saline
(pH 7.2) and then flash frozen in liquid nitrogen. Cell extracts were
analysed using a capillary electrophoresis system supplied by Agi-
lent Technologies (Waldbronn, Germany) equipped with an un-
coated fused silica column (80.5 cm total length, 72 cm effective
length, 75 lm I.D.). Parameters for sample processing and separa-
tion were adapted from those used by Friedecky et al. (2007).
Briefly, the background electrolyte consisted of 40 mM citrate,
0.8 mM cetrimonium bromide (CTAB) adjusted to pH 4.3 with c-
salt: white solid; UV–VIS (phosphate buffer pH 7.0) kmax (log e)
270 (4.312) nm; ESI-MS (pos.) m/z: 454.0 ([M+H]+); ESI-MS (neg.)
m/z: 452.0 ([MꢁH]ꢁ); Mr calculated for the free acid (C17H20
N5O8P): 453.35.
´
butyric acid (GABA). Each new capillary was washed with 1 M
NaOH (30 min) followed by water (30 min) and then running buf-
fer (30 min). At the beginning of every day, the capillary was
washed with solutions in the following order: 1 M NaOH
(10 min), water (10 min) and the running buffer (20 min). Between
each run capillary was washed with the running buffer for 2 min.
Samples were injected under low pressure (50 mbar, 5 s). ATP,
ADP, AMP and oTRMP were identified by spiking with standard
solutions. Quantification was done using corrected peak areas at
detection wavelength 254 nm. The intracellular concentration of
oTR-MP was calculated using the following formula: (concentra-
tion of the analyte in the cell extract ꢂ volume of the cell ex-
tract/number of extracted cells) ꢂ (1/average cell volume). Taking
4.2.4. N6-(2-Hydroxybenzyl)aminopurine riboside-5’-O-diphosphate
(oTRDP)
oTRDP was synthesized from 6-chloropurine-9b-D-riboside-5’-
O-diphosphate, triethylammonium salt, by nucleophilic substitu-
tion with 2-hydroxybenzylamine in the presence of sodium
hydroxide (pH 11) in water at 40 °C. The formation of product
was monitored by analytical HPLC. The reaction was, again,
quenched by addition of formic acid and subsequent cooling to
ꢁ70 °C, then the product was purified by preparative MPLC with
DEAE Sepharose as the stationary phase using 300 mM triethylam-
monium bicarbonate (pH 8) as the mobile phase. The fractions con-
taining product were collected, concentrated under reduced
pressure, and desalted by preparative HPLC (trapping on reversed
phase silica gel, washing with water then eluting with 5% MeOH).
The triethylammonium salt of oTRDP was isolated with a purity
of >99% (yield: 38%). oTRDP triethylammonium salt: white solid;
CEM cells to be spherical, with a diameter of 11.2
lm (the median
value as measured by cell counter Z2, Beckman), the average cell
volume was calculated to be 735.2 l
m3.
UV–VIS (phosphate buffer pH 7.0) kmax (log e) 270 (4.312) nm;
ESI-MS (pos.) m/z: 534.1 ([M+H]+); ESI-MS (neg.) m/z: 532.0
([MꢁH]ꢁ); Mr calculated for the free acid (C17H21N5O11P2): 533.32.
4.4. Cell cultures
The following cell lines – RPMI 8226 (multiple myeloma), CEM
(T-lymphoblastic leukaemia), K562 (chronic myelogenous leukae-
mia), HL-60 (promyelocytic leukaemia), MCF-7 (breast adenocarci-
noma), HeLa (cervical carcinoma), G361 (malignant melanoma),
HOS (human osteosarcoma) and BJ (human foreskin fibroblasts) –
were obtained from the American Type Culture Collection (Manas-
sas, VA, USA). These cells were maintained in standard DMEM or
RPMI medium (Sigma, MO, USA) supplemented with heat-inacti-
4.2.5. N6-(2-Hydroxybenzyl)aminopurine riboside-5’-O-triphosphate,
(oTRTP)
oTRTP was synthesized from 6-chloropurine-9b-D-riboside-5’-
O-triphosphate, triethylammonium salt, by nucleophilic substitu-
tion with 2-hydroxybenzylamine in the presence of sodium
hydroxide (pH 11) in water at 40 °C. Analytical HPLC was used to
monitor the progress of the reaction. The reaction was quenched
by the addition of formic acid and subsequent cooling to ꢁ70 °C,
then the product was purified by preparative MPLC with DEAE Se-
pharose as the stationary phase, using 300 mM triethylammonium
bicarbonate (pH 8) as the mobile phase. The relevant fractions
vated fetal bovine serum (10%) 2 mM L-glutamine and penicillin–
streptomycin (1%) under standard cell culture conditions (37 °C,
5% CO2 in a humid environment) and sub-cultured two or three
times per week using the standard trypsinization procedure.

1358
J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
4.5. Calcein AM cytotoxicity assay
Approximately 10,000 cells in 80
LOX IMVI, OVCAR-3, NCI-H522, U251, UACC-62, MDA-MB-435, OV-
CAR-5 and SF-295. Each mouse was implanted with three cell lines,
as both intraperitoneal (ip) implants and subcutaneous (sc) im-
plants (six implants in total). The compound was administered
intraperitoneally for 4 subsequent days. The activity against the
xenografted cells was assessed by MTT assay. A 50% or greater
reduction in xenograft growth was considered to be a positive re-
sult, and each positive result was given a score of two. The sum of
the scores was then calculated for all implants, giving a total score,
and separate scores for both intraperitoneal (ip) and subcutaneous
(sc) implants for each cell line. Hence, the maximum possible score
was 96 (12 cell lines ꢂ two implantation sites ꢂ two dose lev-
els ꢂ two). A compound is categorised as active if the total score
is at least 20, or the sc score is equal to or greater than eight.
The DTP scoring system has been designed so that standard anti-
cancer drugs are classified as active.
l
l of medium were seeded
into 96-well microtitre plates. After 12 h incubation, compounds
to be tested were added in 20 l portions. Control cultures were
l
treated with DMSO alone. The final concentration of DMSO in the
medium did not exceed 0.5%. Serial, triplicate 3-fold dilutions
(six in total, highest concentration in incubations 166
compound were tested. After 72 h incubation, Calcein AM solution
(Molecular Probes) was added to a final concentration of 1 g/ml,
and the cells were incubated for a further hour. The fluorescence
of free calcein was then quantified using a Fluoroscan Ascent fluo-
rometer (Microsystems), and the percentage of surviving cells in
each well was calculated by dividing the fluorescence obtained
from each cell with exposed cells by the mean fluorescence ob-
tained from control wells ꢂ 100%. Finally, IC50 values (the concen-
trations causing a 50% decrease in cellular esterase activity) were
calculated for each compound from the generated dose–response
curves (Kryštof et al., 2002). The IC50 values presented here are
averages obtained from at least three independent experiments,
where individual replicate values fell within 25% of the average.
lM) of each
l
Acknowledgements
Tests on NCI60, the acute toxicity study and hollow fibre assays
were performed at the Developmental Therapeutics Program, Divi-
sion of Cancer Treatment and Diagnosis, National Cancer Institute
(Bethesda, USA).
4.6. NCI60 cytotoxicity assay
We would like to thank Olga Hustáková and Dita Parobková for
superb technical assistance. We are grateful to Sees-editing Ltd.
(UK) for editing this manuscript to an excellent standard. This
work was supported financially by the Czech Grant Agency (grants
206/06/1284, 301/08/1649) and the Czech Ministry of Education
(grant MSM6198959216).
Tests of toxicity on NCI60, a set of 59 human cancer cell lines de-
rived from nine tissue types, were performed at the Developmental
therapeutics program (DTP) of the National Cancer Institute
(Bethesda, USA). The cytotoxicity of oTR was evaluated by measur-
ing total cell protein using the sulforhodamine B method according
to the standard DTP protocol (http://dtp.nci.nih.gov/docs/compare/
compare_methodology.html) at both time 0 and after 48 h. GI50
values (concentration of a drug inducing 50% reduction of growth)
values were estimated from the dose response curves.
References
Anderson, V.R., Perry, C.M., 2007. Fludarabine: a review of its use in non-Hodgkin’s
lymphoma. Drugs 67, 1633–1655.
Barciszewski, J., Massino, F., Clark, B.F., 2007. Kinetin-a multiactive molecule. Int. J.
Biol. Macromol. 40, 182–192.
4.7. Analysis of the correlation between NCI60 activity and gene
expression patterns
Berge, U., Kristensen, P., Rattan, S.I., 2006. Kinetin-induced differentiation of normal
human keratinocytes undergoing aging in vitro. Ann. NY Acad. Sci. 1067, 332–
336.
Bossi, G., Sacchi, A., 2007. Restoration of wild-type p53 function in human cancer:
relevance for tumor therapy. Head Neck 29, 272–284.
The activity pattern (GI50 values for individual NCI60 cell lines)
of oTR was correlated with those of 214 antineoplastics in the
‘‘Approved Oncology Drugs” and ‘‘Standard Agents” DTP datasets
and with expression patterns (gene expression measurements for
individual NCI60 cell lines) of adenosine kinase (the probes
204119_s_at and 204120_s_at on U133A Affymetrix arrays,
Shankavaram et al., 2007). If a compound had been tested over sev-
eral ranges of concentration, the pattern from the greatest number
of repetitions was used. Pearson correlation coefficients were cal-
culated on a log–log scale. All calculations and manipulations of
data were carried out using statistical software R. The list of ‘‘Stan-
dard Agents” and ‘‘Approved Oncology Drugs” can be accessed at
http://dtp.nci.nih.gov/docs/cancer/searches/standard_agent_table.
html and http://dtp.cancer.gov/branches/dscb/oncology_drugset_
explanation.html, respectively. Relevant activity patterns were
extracted from DTP Cancer Screening Data May 2009 Release
(http://dtp.nci.nih.gov/docs/cancer/cancer_data.html). Microarray
data normalized using the RMA algorithm were downloaded from
the CellMiner database (Shankavaram et al., 2009).
Burnstock, G., 2007. Purine and pyrimidine receptors. Cell Mol. Life Sci. 64, 1471–
1483.
Cabello, C.M., Bair, W.B., Ley, S., Lamore, S.D., Azimian, S., Wondrak, G.T., 2009. The
experimental chemotherapeutic N6-furfuryladenosine (kinetin-riboside)
induces rapid ATP depletion, genotoxic stress, and CDKN1A(p21) upregulation
in human cancer cell lines. Biochem. Pharmacol. 77, 1125–1138.
Carimi, F., Zottini, M., Formentin, E., Terzi, M., Lo, S.F., 2003. Cytokinins: new
apoptotic inducers in plants. Planta 216, 413–421.
Chheda, G.B., Mittelman, A., 1972. N6-(2-isopentenyl)adenosine metabolism in man.
Biochem. Pharmacol. 21, 27–37.
Choi, B.H., Kim, W., Wang, Q.C., Kim, D.C., Tan, S.N., Yong, J.W., Kim, K.T., Yoon, H.S.,
2008. Kinetin riboside preferentially induces apoptosis by modulating Bcl-2
family proteins and caspase-3 in cancer cells. Cancer Lett. 261, 37–45.
Cory, J.G., Mansell, M.M., Whitford, T.W., 1976. Inhibition of ribonucleotide
reductase activity and nucleic acid synthesis in tumor cells by the dialdehyde
derivatives of inosine (NSC 118994) and inosinic acid. Cancer Res. 36, 3166–
3170.
Decker, S., Hollingshead, M., Bonomi, C.A., Carter, J.P., Sausville, E.A., 2004. The
hollow fibre model in cancer drug screening: the NCI experience. Eur. J. Cancer
40, 821–826.
ˇ
Dolezal, K., Popa, I., Hauserová, E., Spíchal, L., Chakrabarty, K., Novák, O., Kryštof, V.,
Voller, J., Holub, J., Strnad, M., 2007. Preparation, biological activity and
endogenous occurrence of N6-benzyladenosines. Bioorg. Med. Chem. 15,
3737–3747.
4.8. Hollow fibre assays
ˇ
Dolezal, K., Popa, I., Kryštof, V., Spíchal, L., Fojtíková, M., Holub, J., Lenobel, R.,
Schmulling, T., Strnad, M., 2006. Preparation and biological activity of 6-
benzylaminopurine derivatives in plants and human cancer cells. Bioorg. Med.
Chem. 14, 875–884.
The effect of two doses (100 and 150 mg/kg/day) of oTR on im-
planted tumours was determined in hollow fibre assays. The doses
selected were derived from acute toxicity experiments that had
been previously carried out (results not shown). The assays were
performed and evaluated at DTP, according to a standard protocol
(http://dtp.cancer.gov/branches/btb/hfa.html). The following cell
lines were used: MDA-MB-231, NCI-H23, SW-620, COLO 205,
Friedecky´, D., Tomková, J., Maier, V., Janoštáková, A., Procházka, M., Adam, T., 2007.
Capillary electrophoretic method for nucleotide analysis in cells: application on
inherited metabolic disorders. Electrophoresis 28, 373–380.
Griffaut, B., Bos, R., Maurizis, J.C., Madelmont, J.C., Ledoigt, G., 2004. Cytotoxic effects
of kinetin riboside on mouse, human and plant tumour cells. Int. J. Biol.
Macromol. 34, 271–275.

J. Voller et al. / Phytochemistry 71 (2010) 1350–1359
1359
Hewett, E.W., Wareing, P.F., 1973. Cytokinins in Populus x robusta (Schneid): light
effects on endogenous levels. Planta 114, 119–129.
Moore, E.C., Loo, T.L., 1984. Inhibition of ribonucleotide reductase by caracemide.
Cancer Treat. Rep. 68, 1293–1294.
Holub, J., Hanuš, J., Hanke, D.E., Strnad, M., 1998. Biological activity of cytokinins
derived from ortho- and meta-hydroxybenzyladenine. Plant Growth Regul. 26,
109–115.
Pieper, R.O., Kennedy, K.A., Mandel, H.G., 1988. DNA-directed actions of 3-
deazaguanine: effects on DNA integrity and DNA elongation/ligation. Cancer
Res. 48, 2774–2778.
Horgan, R., Hewett, E.W., Purse, J.G., Wareing, P.F., 1975. A new cytokinin from
Populus x robusta. Phytochemistry 14, 1005–1008.
Rattan, S.I., Sodagam, L., 2005. Gerontomodulatory and youth-preserving effects of
zeatin on human skin fibroblasts undergoing aging in vitro. Rejuvenation Res. 8,
46–57.
Shankavaram, U.T., Reinhold, W.C., Nishizuka, S., Major, S., Morita, D., Chary, K.K.,
Reimers, M.A., Scherf, U., Kahn, A., Dolginow, D., Cossman, J., Kaldjian, E.P.,
Scudiero, D.A., Petricoin, E., Liotta, L., Lee, J.K., Weinstein, J.N., 2007. Transcript
and protein expression profiles of the NCI-60 cancer cell panel: an integromic
microarray study. Mol. Cancer Ther. 6, 820–832.
Shankavaram, U.T., Varma, S., Kane, D., Sunshine, M., Chary, K.K., Reinhold, W.C.,
Pommier, Y., Weinstein, J.N., 2009. CellMiner: a relational database and query
tool for the NCI-60 cancer cell lines. BMC Genomics 10, 277.
Shoemaker, R.H., 2006. The NCI60 human tumour cell line anticancer drug screen.
Nat. Rev. Cancer 6, 813–823.
Skoog, F., Strong, F.M., Miller, C.O., 1965. Cytokinins. Science 148, 532–533.
Slaugenhaupt, S.A., Mull, J., Leyne, M., Cuajungco, M.P., Gill, S.P., Hims, M.M.,
Quintero, F., Axelrod, F.B., Gusella, J.F., 2004. Rescue of a human mRNA splicing
defect by the plant cytokinin kinetin. Hum. Mol. Genet. 13, 429–436.
Spíchal, L., Rakova, N.Y., Riefler, M., Mizuno, T., Romanov, G.A., Strnad, M.,
Schmülling, T., 2004. Two cytokinin receptors of Arabidopsis thaliana, CRE1/
AHK4 and AHK3, differ in their ligand specificity in a bacterial assay. Plant Cell
Physiol. 45, 1299–1305.
Ikediobi, O.N., Davies, H., Bignell, G., Edkins, S., Stevens, C., O’Meara, S., Santarius, T.,
Avis, T., Barthorpe, S., Brackenbury, L., Buck, G., Butler, A., Clements, J., Cole, J.,
Dicks, E., Forbes, S., Gray, K., Halliday, K., Harrison, R., Hills, K., Hinton, J., Hunter,
C., Jenkinson, A., Jones, D., Kosmidou, V., Lugg, R., Menzies, A., Mironenko, T.,
Parker, A., Perry, J., Raine, K., Richardson, D., Shepherd, R., Small, A., Smith, R.,
Solomon, H., Stephens, P., Teague, J., Tofts, C., Varian, J., Webb, T., West, S.,
Widaa, S., Yates, A., Reinhold, W., Weinstein, J.N., Stratton, M.R., Futreal, P.A.,
Wooster, R., 2006. Mutation analysis of 24 known cancer genes in the NCI-60
cell line set. Mol. Cancer Ther. 5, 2606–2612.
Ishii, Y., Hori, Y., Sakai, S., Honma, Y., 2002. Control of differentiation and apoptosis
of human myeloid leukemia cells by cytokinins and cytokinin nucleosides, plant
redifferentiation-inducing hormones. Cell Growth Differ. 13, 19–26.
Ishii, Y., Sakai, S., Honma, Y., 2003. Cytokinin-induced differentiation of human
myeloid leukemia HL-60 cells is associated with the formation of nucleotides,
but not with incorporation into DNA or RNA. Biochim. Biophys. Acta 1643, 11–
24.
Johnson, J.I., Decker, S., Zaharevitz, D., Rubinstein, L.V., Venditti, J.M., Schepartz, S.,
Kalyandrug, S., Christian, M., Arbuck, S., Hollingshead, M., Sausville, E.A., 2001.
Relationships between drug activity in NCI preclinical in vitro and in vivo
models and early clinical trials. Br. J. Cancer 84, 1424–1431.
Kamath, A., Yoo, D., Stuart, O.A., Bijelic, L., Sugarbaker, P.H., 2009. Rationale for an
intraperitoneal gemcitabine chemotherapy treatment for patients with resected
pancreatic cancer. Recent Pat. Anticancer Drug Discov. 4, 174–179.
Kim, D., Cheng, G.Z., Lindsley, C.W., Yang, H., Cheng, J.Q., 2005. Targeting the
phosphatidylinositol-3 kinase/Akt pathway for the treatment of cancer. Curr.
Opin. Invest. Drugs 6, 1250–1258.
Spinola, M., Colombo, F., Falvella, F.S., Dragani, T.A., 2007. N6-isopentenyladenosine:
a potential therapeutic agent for a variety of epithelial cancers. Int. J. Cancer
120, 2744–2748.
Spinola, M., Galvan, A., Pignatiello, C., Conti, B., Pastorino, U., Nicander, B., Paroni, R.,
Dragani, T.A., 2005. Identification and functional characterization of the
candidate tumor suppressor gene TRIT1 in human lung cancer. Oncogene 24,
5502–5509.
King, M.E., Pfeifle, C.E., Howell, S.B., 1984. Intraperitoneal cytosine arabinoside
therapy in ovarian carcinoma. J. Clin. Oncol. 2, 662–669.
Strnad, M., 1997. The aromatic cytokinins. Physiol. Plant 101, 674–688.
ˇ
Strnad, M., Hanuš, J., Vanek, T., Kamínek, M., Ballantine, J., Fussell, B., Hanke, D.E.,
ˇ
Kryštof, V., Lenobel, R., Havlícek, L., Kuzma, M., Strnad, M., 2002. Synthesis and
1997. Meta-topolin, a highly active aromatic cytokinin from poplar leaves
(Populus x canadensis Moench., cv. Robusta). Phytochemistry 45, 213–218.
Strnad, M., Peters, W., Beck, E., Kamínek, M., 1992. Immunodetection and
identification of N6-(o-hydroxybenzylamino)purine as a naturally occurring
cytokinin in Populus x canadensis Moench cv Robusta leaves. Plant Physiol. 99,
74–80.
biological activity of olomoucine II. Bioorg. Med. Chem. Lett. 12, 3283–3286.
Laezza, C., Caruso, M.G., Gentile, T., Notarnicola, M., Malfitano, A.M., Di, M.T., Messa,
C., Gazzerro, P., Bifulco, M., 2009. N6-isopentenyladenosine inhibits cell
proliferation and induces apoptosis in a human colon cancer cell line DLD1.
Int. J. Cancer 124, 1322–1329.
Laezza, C., Notarnicola, M., Caruso, M.G., Messa, C., Macchia, M., Bertini, S., Minutolo,
F., Portella, G., Fiorentino, L., Stingo, S., Bifulco, M., 2006. N6-
isopentenyladenosine arrests tumor cell proliferation by inhibiting farnesyl
diphosphate synthase and protein prenylation. FASEB J. 20, 412–418.
Letham, D.S., Palni, L.M.S., 1983. The biosynthesis and metabolism of cytokinins.
Annu. Rev. Plant Physiol. 34, 163–197.
Tiedemann, R.E., Mao, X., Shi, C.X., Zhu, Y.X., Palmer, S.E., Sebag, M., Marler, R., Chesi,
M., Fonseca, R., Bergsagel, P.L., Schimmer, A.D., Stewart, A.K., 2008.
Identification of kinetin riboside as a repressor of CCND1 and CCND2 with
preclinical antimyeloma activity. J. Clin. Invest. 118, 1750–1764.
Vazquez, A., Bond, E.E., Levine, A.J., Bond, G.L., 2008. The genetics of the p53
pathway, apoptosis and cancer therapy. Nat. Rev. Drug Discov. 7, 979–
987.
Malspeis, L., Grever, M.R., Staubus, A.E., Young, D., 1990. Pharmacokinetics of 2-F-
ara-A (9-b-
D
-arabinofuranosyl-2-fluoroadenine) in cancer patients during the
Weinstein, J.N., Myers, T.G., O’Connor, P.M., Friend, S.H., Fornace Jr., A.J., Kohn, K.W.,
Fojo, T., Bates, S.E., Rubinstein, L.V., Anderson, N.L., Buolamwini, J.K., van Osdol,
W.W., Monks, A.P., Scudiero, D.A., Sausville, E.A., Zaharevitz, D.W., Bunow, B.,
Viswanadhan, V.N., Johnson, G.S., Wittes, R.E., Paull, K.D., 1997. An information-
intensive approach to the molecular pharmacology of cancer. Science 275, 343–
349.
Weller, M., 1998. Predicting response to cancer chemotherapy: the role of p53. Cell
Tissue Res. 292, 435–445.
Wotring, L.L., Crabtree, G.W., Edwards, N.L., Parks Jr., R.E., Townsend, L.B., 1986.
Mechanism of activation of triciribine phosphate (TCN-P) as a prodrug form of
TCN. Cancer Treat. Rep. 70, 491–497.
Wu, J.J., Weinstein, G.D., Kricorian, G.J., Kormeili, T., McCullough, J.L., 2007. Topical
kinetin 0.1% lotion for improving the signs and symptoms of rosacea. Clin. Exp.
Dermatol. 32, 693–695.
Yegutkin, G.G., Samburski, S.S., Jalkanen, S., 2003. Soluble purine-converting
enzymes circulate in human blood and regulate extracellular ATP level via
counteracting pyrophosphatase and phosphotransfer reactions. FASEB J. 17,
1328–1330.
Yegutkin, G.G., 2008. Nucleotide- and nucleoside-converting ectoenzymes:
important modulators of purinergic signalling cascade. Biochim. Biophys. Acta
1783, 673–694.
phase I clinical investigation of fludarabine phosphate. Semin. Oncol. 17, 18–32.
Meisel, H., Gunther, S., Martin, D., Schlimme, E., 1998. Apoptosis induced by
modified ribonucleosides in human cell culture systems. FEBS Lett. 433, 265–
268.
Mittelman, A., Evans, J.T., Chheda, G.B., 1975. Cytokinins as chemotherapeutic
agents. Ann. NY Acad. Sci. 255, 225–234.
Mlejnek, P., 2001. Caspase inhibition and N6-benzyladenosine-induced apoptosis in
HL-60 cells. J. Cell Biochem. 83, 678–689.
Mlejnek, P., Procházka, S., 2002. Activation of caspase-like proteases and induction
of apoptosis by isopentenyladenosine in tobacco BY-2 cells. Planta 215, 158–
166.
ˇ
Mlejnek, P., Dolezel, P., Procházka, S., 2003. Intracellular phosphorylation of
benzyladenosine is related to apoptosis induction in tobacco BY-2 cells. Plant
Cell Environ. 26, 1723–1735.
Mlejnek, P., Dolezel, P., Procházka, S., 2005. Intracellular conversion of cytokinin
bases into corresponding mononucleotides is related to cell death induction in
tobacco BY-2 cells. Plant Sci. 168, 389–395.
Mlejnek, P., Dolezel, P., 2005. Apoptosis induced by N -substituted derivatives of
ˇ
6
ˇ
adenosine is related to intracellular accumulation of corresponding
mononucleotides in HL-60 cells. Toxicol. In Vitro 19, 985–990.