Biochemical and biological properties of 5-bromotubercidin: Differential effects on cellular DNA-directed and viral RNA-directed RNA synthesis

Article abstract of DOI:10.1016/j.bmc.2007.10.054

We have studied the biochemical and biological properties of 5-bromotubercidin (4-amino-5-bromo-7-β-d-ribofuranosyl-pyrrolo [2,3-d]pyrimidine) (BrTu), a synthetic analogue of the highly cytotoxic pyrrolo[2,3-d]pyrimidine ribonucleoside antibiotic tubercidin (Tu) that interferes with numerous cellular processes, and has been shown to possess biological specificity and selectivity. Thus, BrTu entered the mammalian cell nucleotide pool by phosphorylation, was incorporated into RNA in an unmodified form and, as a consequence, reversibly inhibited (15 μM) mammalian cell growth and the synthesis of high-molecular-weight cellular RNA species (i.e., mRNA and rRNA). However, BrTu (300 μM) did not inhibit picornavirus RNA synthesis or multiplication, and thus discriminated between virus RNA-dependent and all forms of DNA-dependent RNA synthesis whether of cellular or viral origin; because of this BrTu should prove valuable as a metabolic probe for studying the cell-virus relationship. Furthermore, BrTu is a substrate for adenosine kinase (K_m = 24 μM), and is also its potent inhibitor (K_i = 0.93 μM); thus, low concentrations of BrTu (1.5 μM), which did not inhibit cell growth, blocked phosphorylation and the cellular uptake of other, highly cytotoxic pyrrolo-pyrimidine nucleoside analogues (e.g., tubercidin). This block in cellular uptake and incorporation of toxic analogues was associated with the protective effect of BrTu against cell killing by the analogues, providing a mechanism by which BrTu and these analogues can, as we reported elsewhere [J. Virol. 1999, 73, 6444], be used for the selective inactivation of replicating picornaviruses.

Full text of DOI:10.1016/j.bmc.2007.10.054

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 1481–1492
Biochemical and biological properties
of 5-bromotubercidin: Differential effects on cellular
DNA-directed and viral RNA-directed RNA synthesis
Branko Brdara,* and Edward Reichb
aDepartment of Molecular Biology, ‘‘Rudjer Bosˇkovic’’ Institute, Bijenicka 54, 10000 Zagreb, Croatia
bDepartment of Pharmacological Sciences, SUNY at Stony Brook, Stony Brook, NY 11794-8651, USA
Received 28 February 2007; revised 2 October 2007; accepted 17 October 2007
Available online 22 October 2007
Abstract—We have studied the biochemical and biological properties of 5-bromotubercidin (4-amino-5-bromo-7-b-D-ribofuranosyl-
pyrrolo [2,3-d]pyrimidine) (BrTu), a synthetic analogue of the highly cytotoxic pyrrolo[2,3-d]pyrimidine ribonucleoside antibiotic
tubercidin (Tu) that interferes with numerous cellular processes, and has been shown to possess biological specificity and selectivity.
Thus, BrTu entered the mammalian cell nucleotide pool by phosphorylation, was incorporated into RNA in an unmodified form
and, as a consequence, reversibly inhibited (15 lM) mammalian cell growth and the synthesis of high-molecular-weight cellular
RNA species (i.e., mRNA and rRNA). However, BrTu (300 lM) did not inhibit picornavirus RNA synthesis or multiplication,
and thus discriminated between virus RNA-dependent and all forms of DNA-dependent RNA synthesis whether of cellular or viral
origin; because of this BrTu should prove valuable as a metabolic probe for studying the cell–virus relationship. Furthermore, BrTu
is a substrate for adenosine kinase (Km = 24 lM), and is also its potent inhibitor (Ki = 0.93 lM); thus, low concentrations of BrTu
(1.5 lM), which did not inhibit cell growth, blocked phosphorylation and the cellular uptake of other, highly cytotoxic pyrrolo-
pyrimidine nucleoside analogues (e.g., tubercidin). This block in cellular uptake and incorporation of toxic analogues was associated
with the protective effect of BrTu against cell killing by the analogues, providing a mechanism by which BrTu and these analogues
can, as we reported elsewhere [J. Virol. 1999, 73, 6444], be used for the selective inactivation of replicating picornaviruses.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
since it corresponds to the location of the N-7 of ade-
nine; and the presence at this position of even bulky
space filling groups does not obstruct the H-bonding
sites involved in base-pairing, or otherwise interfere with
the formation of polynucleotide helices.5–8 Thus, pyrrolo-
[2,3-d]pyrimidine nucleotides are efficient substrates for
polymerizing enzymes and may form polymers with
unusual physical and biochemical properties. Finally,
the large number of derivatives, substituted with a vari-
ety of functional groups, may provide metabolic probes
which possess biological selectivity and specificity with
respect to nucleotide and polynucleotide function. For
these reasons we have studied some of the biochemical
and biological properties of 5-bromotubercidin (4-ami-
no-5-bromo-7-b-D-ribofuranosyl-pyrrolo[2,3-d]pyrimidine)
(BrTu) (Fig. 1), a synthetic derivative of pyrrolo[2,3-
d]pyrimidine ribonucleoside antibiotic tubercidin1 (Tu);
the aglycone of BrTu (4-amino-5-bromo-pyrrolo[2,3-d]
pyrimidine) is the natural product isolated from a mar-
ine sponge.9 Tubercidin is a highly cytotoxic adenosine
analogue that interferes with numerous cellular pro-
Previous studies of pyrrolo[2,3-d]pyrimidine nucleosides
and nucleotides have demonstrated that these purine
analogues are incorporated into nucleic acids by mam-
malian cells in culture1–4 and by enzyme preparations
in vitro.5,6 Furthermore, these compounds can substi-
tute for the normal adenosine counterparts in a wide
variety of enzyme reactions. In comparison with ade-
nine, the pyrrolo[2,3-d]pyrimidine structure provides
an extra valence which permits the introduction of di-
verse functional groups, thereby making possible the
preparation of a wide variety of ring-substituted deriva-
tives. The position of the extra valence is significant,
Keywords: 5-Bromotubercidin; Pyrrolo[2,3-d]pyrimidine nucleoside
analogues; Metabolic probe; RNA-dependent RNA synthesis; Picor-
naviruses; Mouse fibroblasts; Adenosine kinase inhibitor; Selective
antiviral activity; RNA viruses; DNA viruses.
*
Corresponding author. Tel.: +385 1 4561093; fax: +385 1 4561177;
e-mail: brdar@rudjer.irb.hr
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.10.054

1482
B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
possesses biological selectivity and specificity with re-
spect to nucleotide and polynucleotide function. We
therefore investigated the substrate specificity of BrTu
and its phosphorylated derivatives for a variety of cellu-
lar, viral and enzymatic reactions. Several aspects of
BrTu action were important: (1) BrTu inhibited the
growth of mouse fibroblasts in culture (L2-cells) at rela-
tively low concentrations (Fig. 2); complete growth sto-
page required 15 lM BrTu. At a BrTu concentration of
3 lM, a consistent rate of cell multiplication was main-
tained, although it was not quite as rapid as in untreated
control cultures. When cells which had been exposed to
BrTu (15–300 lM) were washed and incubated in fresh
medium free of inhibitor, growth resumed relatively rap-
idly. The lag period which intervened between the re-
moval of BrTu and resumption of growth increased
with increasing time of exposure to a single drug concen-
tration (30 lM) (Fig. 3a), but it did not change with
increasing drug concentrations (15–300 lM) at a fixed
time (8 h) of exposure (Fig. 3b). Continued microscopic
examination of the cultures revealed that the increase in
cell number which took place after BrTu removal was
not due to excessively rapid growth of a minority of
the cell population: all the cells seen in many fields under
microscopic observation divided mitotically in a normal
manner. In contrast to BrTu, whose effects on cell
growth are reversibly cytostatic, its parent compound
Tu is both more potent and, at concentrations above
37 nM, rapidly and irreversibly cytotoxic (data not
shown and Ref. 10). (2) The data in Figure 4 show that
Figure 1. Structures of 5-bromotubercidin and the related analogues,
tubercidin and deazanebularin, used in the present work.
cesses including mitochondrial respiration, de novo pur-
ine synthesis, rRNA processing, methylation of tRNA
and nucleic acid and protein synthesis.1,4 The results
presented here reveal that in contrast to the highly cyto-
toxic nucleoside tubercidin, the C-5 substituted ana-
logue BrTu is more selective in its biological activity
involving a variety of cellular and enzymatic reactions.
For instance, BrTu reversibly inhibits the synthesis of
cellular RNA, but does not inhibit either picornavirus
RNA synthesis or multiplication, providing a means
for discriminating between viral and host replicating sys-
tems. Furthermore, by inhibiting adenosine kinase,
BrTu strongly inhibits the phosphorylation and cellular
uptake of other, highly cytotoxic pyrrolo-pyrimidine
ribonucleosides, such as Tu and deazanebularin (DN)
(Fig. 1). This block in cellular uptake of Tu or DN
was associated with the protective effect of BrTu against
cell killing by the analogues, providing the mechanism
by which BrTu and these analogues can be used for
the selective inactivation of replicating picornaviruses.10
2. Results and discussion
2.1. 5-Bromotubercidin inhibits cell growth and macro-
molecule synthesis reversibly
Our previous studies of antiviral properties of a series of
pyrrolo-pyrimidine nucleoside analogues of adenosine,
including tubercidin,1 deazanebularin,2 toyocamycine,11
sangivamycin, formycin and 5-bromotubercidin,10 have
shown that only 5-bromotubercidin (BrTu) possesses
unusual selectivity with respect to its effects on some
RNA and DNA containing viruses in cultured cells.
Hence, BrTu was totally inert with regard to picornavi-
rus (e.g., mengovirus) RNA synthesis or multiplication,
whereas it strongly suppressed the synthesis of RNA in
host cells just as it did the growth of DNA viruses (e.g.,
vaccinia virus). This differential effect of BrTu on the
cellular DNA-directed and picornavirus RNA-directed
RNA synthesis was of interest for two reasons. First,
it suggested that BrTu can be used as an experimental
model system for studying the cell–virus relationship
in the background of cellular RNA synthesis inhibition,
and second that it may provide a metabolic probe which
Figure 2. Effect of 5-bromotubercidin on the growth of L-cells. Cells
were plated at 1 · 105 per 60 mm Petri dish and incubated overnight.
BrTu was then added and maintained at the indicated concentrations
throughout the period of observation. Cell counts were performed at
the times shown on defined areas of duplicate plates. (s), control; (h),
BrTu (3 lM); (·), BrTu (15 lM); (d), BrTu (30 lM).

B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
1483
concentration of BrTu (to 300 lM) did not produce
any additional inhibition of RNA synthesis. These ef-
fects on RNA and DNA synthesis, like those on growth,
were reversed on removal of BrTu (data not shown).
The time required for restoration of the normal rate of
RNA and DNA synthesis was inversely related to period
of exposure to BrTu. (3) The results of several experi-
ments demonstrated that the inhibition of cellular
RNA synthesis by BrTu mainly involved the species of
ribosomal RNA, and that the synthesis of both rapidly
labelled nuclear 45 S RNA and tRNA was more resis-
tant to the drug (Fig. 5); the synthesis of heterogeneous
nuclear RNA was, as determined previously,12 also
inhibited by BrTu (data not shown). Cultures treated
by BrTu synthesized as compared with untreated con-
trols approximately 5-fold less amount of rapidly la-
belled 45 S RNA (Fig. 6), whose conversion to rRNA
species was apparently impaired in the presence of the
drug (Fig. 7); the rapidly labelled 45 S RNA, which
was synthesized in the presence of BrTu, was following
drug removal not normally processed into ribosomal
RNA species in the presence of actinomycin D (AMD)
(Fig. 8, Chart II + AMD), as compared with untreated
controls (Chart I + AMD; Chart III ꢀ AMD). (4) Sev-
eral lines of experiments have been performed to test
whether the inhibition of RNA synthesis by BrTu re-
flects some interference with polymerization or is it
due to changes in the metabolism of nucleotide mono-
mers. To test for a possible effect of BrTu on purine
metabolism, experiments were performed with [14C]for-
mate as radioactive precursor for polynucleotide synthe-
sis. If BrTu inhibited nucleic acid synthesis by blocking
the formation of purine nucleotides, the incorporation
of formate into RNA and DNA should be more sensi-
tive than that of nucleosides and aglycons. However,
this was not the case: a given concentration of BrTu
inhibited the incorporation of formate into RNA to ex-
actly the same degree as that of nucleoside precursors
(data not shown). One corollary of this result is that
the incorporation of formate into acid-soluble nucleo-
tide pool was also unaffected by BrTu. This suggested
that unlike many purine and purine nucleoside ana-
logues13 including tubercidin,4 BrTu did not inhibit pur-
ine biosynthesis de novo. Therefore, its inhibitory effects
on cell growth and macromolecule synthesis are proba-
bly the result of a primary action of BrTu on polynucle-
otide synthesis or metabolism.
Figure 3. 5-Bromotubercidin inhibits cell growth reversibly. The
conditions for cell growth were as in Figure 2. Cell cultures were
incubated either for various lengths of time with a single drug
concentration (30 lM) (a), or for 8 h with indicated concentrations of
BrTu (b). At the indicated times thereafter the cultures were washed
and incubation continued in drug-free medium. Cell counts were
performed at the indicated times on defined areas of duplicate plates.
(a) Control (s); BrTu (30 lM): (·), 2 h; (d), 4 h; (h), 8 h; (n), 14 h;
(j), continuous. (b) Control (s); BrTu: (·), 15 lM; (d), 30 lM; (h),
60 lM; (n), 150 lM; (m), 300 lM; (j), 15 lM continuous.
2.2. Metabolism of 5-bromotubercidin: characterization
of acid-soluble nucleotides and incorporation into nucleic
acid
of the classes of macromolecules examined, BrTu prefer-
entially inhibited the synthesis of both RNA (4A) and
DNA (4B); the synthesis of protein was relatively unim-
paired for some hours (data not shown). DNA synthesis
was gradually slowed with increasing concentrations of
BrTu (from 15 to 300 lM) and was ultimately com-
pletely arrested. RNA synthesis was also slowed by
BrTu (4A). The degree of inhibition rose rapidly over
the concentration range of 3–15 lM and reached a max-
imum at 15–30 lM. At these concentrations RNA syn-
thesis was reduced to 15–20% of the normal and may
proceed at this rate for up to 9 h. Further increases in
In view of the above observations it is desirable to ob-
tain information concerning the metabolism of BrTu
and the incorporation, if any, into polynucleotides. As
seen in Table 1 the fraction of radioactive BrTu in the
culture medium which appeared in the cellular nucleo-
tide pool was very small; the maximum concentration
was reached within 10–25 min and did not change mea-
surably during several hours. The distribution of acid-
soluble radioactivity resembled that found for other
metabolizable nucleoside analogues in that most of the
residues (>80%) were in the form of the 50-triphos-

1484
B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
Figure 4. Effects of 5-bromotubercidin on macromolecule synthesis in L-cells. Cell monolayers (3 · 105 per 60 mm Petri dish) were exposed to
indicated concentrations of BrTu followed 20 min later by [3H]guanosine (2 lCi/ml) for DNA and RNA synthesis. At the indicated times the cells
were washed, and the radioactivity incorporated into RNA (a) and DNA (b) was measured as described in Section 3. (s), control; BrTu: (·), 3 lM;
(j), 15 lM; (d), 30 lM; (h), 60 lM; (n), 150 lM; (m), 300 lM.
Figure 5. Synthesis of RNA in nuclear, ribosomal and postribosomal fractions of L-cells treated with 5-bromotubercidin. Cell monolayers (3 · 106
per 100 mm Petri dish, 4 dishes) were exposed to BrTu (37.5 lM; 12.5 lg/ml), followed 20 min later by [3H]guanosine (2 lCi/ml), and incubation was
continued for 2 h. Cells were then collected, disrupted in a homogenizer, and RNA was analysed by sucrose density gradient (20–50%) centrifugation,
after being extracted from nuclear, ribosomal and postribosomal fractions (Section 3). (s), A260 nm; (·), nuclear RNA; (d), ribosomal RNA; (n),
soluble RNA.
phates. The flow of [3H]BrTu into acid-insoluble cellular
fractions shows that at 24 lM BrTu the radioactivity
incorporated into RNA reached the maximal value
within one hour after addition of the drug (Fig. 9).
When the cells were incubated in the presence of
2.4 lM BrTu the incorporation of radioactive com-
pound into RNA was not completely interrupted and
proceeds beyond the level which was observed in the cul-
ture treated with 24 lM BrTu. Incorporation of BrTu
into cellular RNA occurred through DNA-dependent
reaction, since it was sensitive to inhibition by actinomy-
cin D (data not shown). BrTu was also incorporated
into DNA, but the total amount of radioactivity found
in this macromolecule was also quite low; we did not
analyse whether BrTu was transformed to a 20-deoxy
derivative intracellularly and incorporated in this form
into DNA. A low but measurable amount of radioactive
BrTu, which was incorporated into cellular RNA, was
found to be in an unmodified form. Thus, the radioac-
tive material released on alkaline hydrolysis of RNA,

B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
1485
Figure 6. Effect of 5-bromotubercidin on the synthesis of RNA species in nucleolar fraction. Experimental conditions were as in Figure 5, except that
the cells were incubated with radioactive guanosine for 30 min. After this, the cells were collected and separated into nucleolar and nucleoplasmic
fractions, and the RNA was extracted and analysed on polyacrylamide gel electrophoresis (Section 3).
Figure 7. Effect of 5-bromotubercidin on the synthesis of RNA species in nucleoplasmic fraction. Experimental conditions were as in Figure 6.
which was isolated from tRNA fraction, was chromato-
graphically and electrophoretically indistinguishable
from 20(30)BrTuMP. It was converted by Escherichia coli
alkaline phosphatase to a compound which chromato-
graphically resembled authentic BrTu (data not shown).
The radioactivity incorporated into tRNA was found to
be in the nucleotide (81%) and nucleoside (19%) form.
This suggests that, apart from phosphorylation, BrTu
does not undergo metabolic modification in fibroblast
cultures.
Several of the above observations are unexpected. In
particular, the fact that the uptake of BrTu reaches a
plateau quickly, corresponding to a very low concentra-
tion in the nucleotide pool, is in marked contrast to the
phosphorylation of adenosine and all other adenosine

1486
B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
Figure 8. 5-Bromotubercidin affects the maturation of ribosomal 45 S RNA precursor. Three sets of cultures were established (3 · 106 cells per 100
mm Petri dish, 4 dishes), one-set receiving BrTu (37.5 lM; 12.5 lg/ml) (II), and two sets serving as untreated controls (I and III). After 20 min all sets
were labelled by a 20 min pulse of [3H]uridine (2 lCi/ml; spec. act. 5 Ci/mmol), washed and incubated for additional 4 h in fresh drug-free medium
(III) containing 0.1 lg/ml of actinomycin D (I and II). At the end of the period of incubation the nucleoli were isolated, RNA extracted and analysed
as in Figure 6. I, control (+AMD); II, BrTu (+AMD); III, control (ꢀAMD).
analogues examined to date.1,2,10,12 This low but mea-
Table 1. Cellular uptake of [3H]5-bromotubercidin into acid-soluble
surable incorporation of BrTu nucleotides into RNA
pool
coupled to BrTu mediated block in the synthesis and
[3H]5-Bromotubercidin
Incubation
(min)
Radioactivity
(cpm)
maturation of high-molecular-weight cellular RNA
(Figs. 7 and 8), probably accounts for the excellent
recovery of cells from the treatment with BrTu. Further,
the low cytotoxicity of BrTu, its ready reversibility and
inertness to purine synthesis de novo, all suggested that
the uptake of BrTu is not limited by generalized break-
down of cellular metabolism or any significant change in
nucleotide metabolism. We presumed that the cellular
uptake of BrTu might be impaired at the level of the
cytosolic nucleoside or nucleotide phosphorylation
and/or membrane nucleoside transportation. We have
performed a series of experiments to identify the reac-
tion which restricts BrTu uptake. For this purpose we
have explored the phosphorylation of BrTu and of some
(lM)
2.4
2.4
2.4
24
25
120
240
25
481
459
509
825
767
794
24
24
120
240
Cell cultures (2 · 105 cells per 60 mm Petri dish) were incubated for
various lengths of time with indicated concentrations of [3H]BrTu
(spec. act. 4.8 lCi/lmol). At the indicated times, the cells were thor-
oughly washed with phosphate-buffered saline, the nucleotide pool
extracted with 0.25 M perchloric acid for 45 min in an ice bath and the
radioactivity determined (Section 3).
Figure 9. Incorporation of [3H]5-bromotubercidin into DNA and RNA. The conditions for cell growth (8 · 105 cells per 60 mm Petri dish) and
labelling with radioactive BrTu were as in Table 1. The cells remaining after extraction with 0.25 M perchloric acid were resuspended in alkali, and
the radioactivity in RNA and DNA was determined (Section 3). BrTu: 24 lM, (·); 2.4 lM, (s).

B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
Table 2. Nucleoside and nucleotide kinase activity in L-cell homogenate or S100 fraction
1487
Compound
Fraction of nucleoside converted to phosphatesa (cpm)
Crude cell homogenateb S100 fractionb
Monophosphate
Triphosphate
Monophosphate
Triphosphate
Bromotubercidin
Deazanebularin
Adenosine
0.084 0.011
0.53 0.22
0.35 0.095
0.010 0.002
0.050 0.002
0.135 0.041
0.078 0.023
0.640 0.170
0.142 0.053
0
0.012 0.003
0.470 0.154
a Reaction products after electrophoretic separation of radioactive nucleotides (Section 3).
b Average SD for three separate experiments.
Table 3. Kinetic constants for nucleoside analogues with rabbit liver adenosine kinase
a
Compound rel. Vmax/Km
Km (lM)
Kib (lM)
Vmax (lM minꢀ1
)
Relativec Vmax
Substrate efficiency
Adenosine
BrTu
DN
5.5
24
3.13
2.94
3.0
100
94
16.95
3.77
7.7
0.93 (Adn); 0.27 (DN); 0.5 (Tu)
12.5
9.5
96
114
Tu
3.57
12.0
a Km and Vmax values were derived from five velocity measurements over a 10-fold nucleoside concentration which spanned the Km value.
b The K values for inhibition of the enzyme by BrTu for adenosine, DN and Tu were 0.93, 0.27 and 0.5 lM, respectively.
c Vmax values are relative to the Vmax of adenosine which was arbitrarily set at 100.
other adenosine analogues, such as tubercidin (Tu) and
deazanebularin (DN), in crude and in partially fraction-
ated homogenates of L-cells. From the data presented in
Tables 2 and 3, the following facts may be noted: (1) All
the nucleosides tested (BrTu, DN, adenosine) and their
corresponding 50-monophosphates were converted to
their 50-triphosphates in crude homogenates of L-cells
(Table 2). BrTu was, however, in this reaction less effec-
tive substrate than adenosine or DN; this matches our
observation (Fig. 10) that BrTu was a substrate for par-
tially purified rabbit liver adenosine kinase
(Km = 24 lM), although less effective than adenosine
(Km = 5.5 lM). (2) Only adenosine and its 50-mono-
phosphate, and to a lesser extent DN and its 50-mono-
phosphate, were phosphorylated to their triphosphates
in the 100,000g supernatant from such homogenates
(Table 1). BrTu and its 50-monophosphate were there-
fore converted to the triphosphate only in fractions
which contain some of the particulate fractions of L-
cells. (3) The presence of BrTu strongly inhibited the
phosphorylation of DN and Tu both in cell homogenate
and rabbit liver adenosine kinase preparations; the
phosphorylation of adenosine in the respective enzyme
preparations was moderately inhibited by BrTu. Thus,
the K values for inhibition of the rabbit liver adenosine
kinase by BrTu for adenosine, DN and Tu were 0.93,
0.27 and 0.5 lM, respectively (Table 3). Replacement
of N-7 in adenosine by C–Br, as occurs in BrTu, did
not appreciably affect the value of Vmax while causing
a 5-fold increase in the Km value, thus having a marked
effect upon the substrate efficiency of BrTu. A compar-
ison of BrTu with tubercidin and deazanebularin exhib-
ited a slight increase in the maximal velocity but showed
a 2- to 2.5-fold decrease in the Km values, thus making
BrTu inferior to these compounds in terms of substrate
efficiency.
(4)
Bromotubercidin
50-triphosphate
(BrTuTP) was shown to be an effective energy source
for the phosphorylation of adenosine and nucleoside
analogues or of BrTu itself (data not shown). In addi-
tion, BrTuTP was a substrate for E. coli RNA polymer-
ase with the synthetic polydeoxyribonucleotides as
templates (data not shown). Efficiency of BrTuTP to
polymerize in these reactions is comparable to that of
rATP, indicating that the bulky C–Br group does not
interfere with the formation of BrTu containing polynu-
cleotides. These observations are of some interest for
two reasons. First, they show that BrTu is a substrate
for adenosine kinase, and a second result of significance
is that BrTu inhibits the phosphorylation of other aden-
osine analogues, such as deazanebularin and tubercidin.
Since these analogues are rapidly and irreversibly cyto-
toxic, whereas BrTu is not, it was possible to test
whether the observations at the level of cell-free homog-
Figure 10. Lineweaver-Burk plots of rabbit liver adenosine kinase
activity at four concentrations of BrTu (·) in the range 10–120 lM,
and at five concentrations of adenosine (s) in the range 5–50 lM. The
plotted data were from a single experiment in triplicate.

1488
B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
Table 4. BrTu blocks cellular uptake of tubercidin
adenosine kinase involving the intermediate formation
of analogue base.17 On the other hand, multiple path-
ways are available for the corresponding adenine nucle-
otides including the de novo purine nucleotide
biosynthetic pathway, which is not affected by BrTu.
Since the nucleoside uptake is a result of both membrane
transport and subsequent intracellular metabolism, it is
likely that BrTu inhibits nucleoside analogue (e.g., Tu or
DN) uptake at the level of the membrane nucleoside
transporter coupled to inhibition of adenosine kinase;
the degree of inhibition of analogue uptake by BrTu will
therefore be determined in part by the analogue’s ability
to cross the membrane as a substrate for the nucleoside
transporter. Several lines of evidence indicate the exis-
tence of two major transporter families that mediate
nucleoside permeation of mammalian cells: the concen-
trative nucleoside transporters are Na+-dependent, and
the equilibrative nucleoside transporters are Na+-inde-
pendent.15,18 It was recently reported19 that 5-iodotub-
ercidin, a potent adenosine kinase inhibitor just like
BrTu, showed a significant inhibition of adenosine up-
take and adenine nucleotide formation at concentrations
above 100 nM in rat glioma cells expressing an equili-
brative transporter, while concentrations above 3 lM
were required for cells expressing concentrative trans-
porter. This study demonstrates that the effects of 5-iod-
otubercidin in whole cell assays are dependent upon
nucleoside transporter expression independently of its
inhibition of adenosine kinase. Thus, cellular and tissue
differences in expression of nucleoside transporter sub-
types may affect the pharmacological action of adeno-
sine kinase inhibitors including BrTu. It may therefore
be possible to promote selectively, in the presence of
BrTu, the incorporation of the toxic analogue (e.g., Tu
or DN) in particular target tissues, while protecting spe-
cific tissues.
Incubation with
[3H] tubercidin (h)
Radioactivity (% inhibition)
Acid-soluble pool
RNA
DNA
1
2
3
94.3
96.9
97.6
94.4
95.6
95.4
89.3
94.2
94.8
Two parallel sets of cultures were established (6 · 105 cells per 60 mm
Petri dish), one-set receiving BrTu (30 lM). After 30 min both sets
were supplemented with [3H]tubercidin (3.7 lM; spec. act. 155 mCi/
mmol). Dishes were withdrawn from each set of cultures at the indi-
cated times, washed and analysed for radioactivity in the acid-soluble
pool, RNA and DNA (Section 3).
enates could be reproduced with the intact cells. Thus,
the incorporation of [3H]Tu (Table 4) or [3H]DN (data
not shown) into acid-soluble and acid-insoluble material
of L-cells was reduced to almost undetectable levels in
the presence of BrTu. Electrophoretic analysis of acid-
soluble pool revealed that Tu was not phosphorylated
in the presence of BrTu (data not shown), whereas in
control cultures most of the radioactivity was found in
the triphosphate region.1,2 Interestingly, the BrTu med-
iated block in cellular uptake and incorporation of toxic
Tu and/or DN was associated with the protective effect
of BrTu against cell killing by the toxic analogues. Thus,
the growth of BrTu-treated cell cultures, even those
simultaneously exposed to DN, was following drug re-
moval the same as that of the untreated control, whereas
cell multiplication did not resume in the culture that had
been transiently incubated with DN alone (Table 5).
Identical results were obtained with Tu in comparable
experiments.10 In short, the presence of BrTu simply
nullified the cytotoxic action of other adenosine ana-
logues. This suggested that the extent of cellular uptake
and incorporation of a given toxic analogue could be
modulated by BrTu. The possibility of controlling the
rate of analogue metabolism in this way encourages at-
tempts to differentiate between viral and host replicating
systems10 and possibly between different cell and tissue
types for the rational design of pharmacologically active
nucleoside analogues.
2.3. Effects of 5-bromotubercidin on growth of some
animal viruses
The growth of viruses in cultured cells provides an excel-
lent system for detecting the differential susceptibility to
inhibitors of the various nucleic acid replication reac-
tions. We have studied the effects of BrTu on the life cy-
cles of several RNA viruses (mengovirus, poliovirus,
reovirus, vesicular stomatitis virus, Newcastle disease
virus, sindbis virus, influenza virus, Rous sarcoma
virus), and three DNA viruses (vaccinia virus, SV40,
herpes simplex virus). The results of these experiments
were described in detail elsewhere.10,12,20,21 The observa-
The rationale for achieving differential effects of ana-
logues could be as follows. Adenosine analogues (BrTu,
Tu, DN) can enter the nucleoside pool either via the
nucleoside transport system14,15 driven by rapid intracel-
lular phosphorylation by cytosolic nucleoside and nucle-
otide kinases,16 or by the pathways independent of
Table 5. BrTu protects against the cytotoxic action of DN
Incubation with nucleoside analogue (h)
Relative cell count
Control
BrTu (60 lM)
alone
BrTu + DN
(8 lM)
BrTu + DN
(20 lM)
BrTu + DN
(40 lM)
DN (40 lM)
alone
18
28
45
1.96
3.53
8.20
1.72
3.50
6.80
1.66
3.32
7.07
1.93
3.30
8.00
1.72
3.32
7.35
0.75
0.48
0
L-cell monolayers (8 · 104 per 60 mm Petri dish) were incubated for 5 h with the indicated concentrations of each nucleoside, where BrTu was present
60 lM. After 5 h, the cultures were washed and incubation was continued in drug-free medium. Cells were counted at the indicated times in defined
areas of each plate. The presence of BrTu protected the cells against the toxicity resulting from the transient exposure to DN.

B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
Table 6. Effects of 5-bromotubercidin on virus multiplication
1489
completely resistant to concentrations of BrTu
(300 lM) much higher than those (15 lM) which
strongly inhibited the growth of the host cell (Fig. 2),
of vaccinia virus (Table 6), and of the DNA-dependent
RNA synthesis of uninfected cells (Fig. 4a). These find-
ings are of interest for several reasons: (1) It is apparent
that BrTu inhibits cellular DNA-directed RNA synthe-
sis, and, by implication, also that of the DNA virus vac-
cinia; in contrast the growth and RNA synthesis of
mengovirus are unaffected. This distinction by BrTu be-
tween DNA-dependent and RNA-dependent RNA syn-
thesis is not the result of differences in cytological
localization of the two processes, since the entire life cy-
cle of vaccinia virus like that of mengovirus runs its
course in the cytoplasm; BrTu also inhibits the growth
of SV4021 or herpes simplex virus both of which repli-
cate in the nucleus of the respective host cells. The differ-
ential susceptibility of the two replication reactions
more likely reflects differences in the interaction of
BrTuTP with the respective polymerizing systems. In-
deed, experiments using [3H]BrTu showed that BrTu is
not incorporated into mengovirus RNA or virus parti-
cles (data not shown), which could be the reason for
the inertness of BrTu in both the growth and RNA syn-
thesis of mengovirus. Since BrTuTP is an in vitro sub-
strate for both mengovirus RNA polymerase22 and
E. coli RNA polymerase (data not shown), and thus
substitutes completely for ATP in pairing with uracil,
it is possible that substrate selection by mengovirus
polymerase is not determined solely by the constraints
of suitable base-pairing. Additional determinants are
therefore involved in substrate selection: (i) Given the
fact that the fraction of radioactive BrTu in the culture
medium of both uninfected and mengovirus-infected
cells, which appeared in cellular nucleotide pool (Table
1) and cellular RNA (Fig. 9), was very small but unde-
tectable in mengovirus RNA (data not shown), it is pos-
sible that low levels of BrTu nucleotide pool coupled to
the previously reported22–24 low substrate affinity (Km)
of mengovirus RNA polymerase for nucleoside ana-
logue triphosphates might account for the lack of BrTu
incorporation into viral RNA. Namely, replacement of
the natural rATP by BrTuTP (as shown in Ref. 22) un-
iquely alters the kinetics of the mengovirus polymerase
reaction by sharply reducing initial rates of synthesis
in non-linear fashion. (ii) Since mengovirus plus-
stranded RNA (which serves as messenger RNA) repli-
cates through a minus-stranded RNA intermediate,
these observations suggest that mengovirus replicase
has a unique ability to respond to its plus and minus
strands. Thus, the initial product of plus strand-primed
replicase reaction in the presence of substrate analogue
(BrTuTP) would be minus strands containing analogue
base which is not properly recognized by the mengovirus
replicase as a biologically significant template. (iii) Addi-
tional reason for this differential incorporation of BrTu
might be that viral and cellular polymerases draw on dif-
ferent intracellular nucleotide pools.25
Virus
5-Bromotubercidin Incubation No. of PFU/cell
(lM)
(h)
Mengovirus
None
None
15
0
2
24
24
24
24
24
24
560
868
875
806
953
706
30
75
150
300
Vaccinia virus None
0
0.10
79
None
15
30
24
24
24
24
24
24
18
11.5
7.4
75
150
300
2.0
0.55
Monolayers of L2-cells (2 · 105 cells per 60 mm Petri dish) were
incubated for 20 min with indicated concentrations of BrTu after
which the cells, in separate plates, were infected with mengovirus and
vaccinia virus at a multiplicity of 10 plaque-forming units (PFU)/cell,
each. At the end of the period of adsorption (1 h at 37 ꢁC) the plates
were washed three times with culture medium and one plate for each
virus type was frozen as zero-time control. Other dishes received 3 ml
of complete medium with BrTu as indicated. After 24 h of incubation
all cultures were frozen and thawed three times and virus titrations
were performed as in Ref. 10.
tions which are relevant for the present discussion are gi-
ven in Table 6 and Figure 11. It can be seen that both
the growth and the synthesis of mengoviral RNA were
Figure 11. 5-Bromotubercidin does not block mengovirus RNA
synthesis. L-cells were plated at 3 · 105 per 60 mm Petri dish and
incubated overnight. Actinomycin D (2 lg/ml) was added to all
cultures and maintained thereafter throughout the experiment. Twenty
minutes later mengovirus (10 PFU/cell) was added in a small volume
(0.3 ml) and adsorbed to the monolayer during 1 h at 37 ꢁC; one set of
cultures was mock infected. Two sets of cultures then received medium
containing BrTu as indicated, the other set serving as control, and
incubation was continued for 20 min, at which time [3H]uridine (2 lCi/
ml) was added to all plates. At the indicated times thereafter the
radioactivity incorporated into material insoluble in 5% trichloroacetic
acid was measured as described in Section 3. Control (s); BrTu:
15 lM (n); 60 lM (h); mock (d).
(2) Another noteworthy conclusion can be drawn from
the complete resistance to BrTu of mengovirus growth.
Since the normal growth of the virus requires the main-
tenance of the integrity of cellular systems for protein

1490
B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
synthesis and energy metabolism, it follows that BrTu
does not interfere with these complex processes. Taken
together, our results suggest that BrTu produces a single
pattern of effects on picornavirus RNA and all forms of
DNA transcription, whether of viral or cellular origin.
This distinction by BrTu between cellular DNA-depen-
dent and picornavirus RNA-dependent RNA synthesis
could be used for designing specifically antiviral nucleo-
sides, which we reported elsewhere.10 Briefly, being a po-
tent adenosine kinase inhibitor BrTu can be used to
prevent cellular uptake of other, more cytotoxic adeno-
sine analogues, such as Tu (Table 4) and DN. This block
in cellular uptake of Tu is associated with the protective
effect of BrTu against cell killing by the toxic analogue.
However, in mengovirus-infected cells BrTu does not
prevent Tu from incorporating into mengovirus RNA,
giving rise to functionally defective polynucleotides that
abort the virus growth cycle under conditions that do no
affect the viability of normal uninfected cells.
nucleotides, amino acids and other metabolites, radioac-
tive and non-radioactive, were obtained from regular
commercial sources, and were the best grade available.
Radioactive 5-bromotubercidin, tubercidin and deazan-
ebularin were prepared by New England Nuclear Cor-
poration, Boston, Mass, through catalytic tritium
exchange in dimethylformamide. The crude radioactive
product was purified to a single chromatographically
homogeneous material of constant specific radioactivity
by paper and column chromatography as previously
described.2,6
Mono-, di- and triphosphates were chemically synthe-
sized as described by Ward et al.6
3.2. Synthesis of poly r(BrTu-U) and poly r(BrTu)
The conditions for synthesizing BrTu riboheteropoly-
mers and ribohomopolymers were previously de-
scribed.6 Briefly, one millilitre reaction mixture
contained 40 mM Tris–HCl, pH 7.9, 1 mM MnCl2,
4 mM MgCl2, 3 mM 2-mercaptoethanol, 0.04 mM poly
d(A-T) or 0.09 mM poly (dT), 0.4 mM [3H]rATP or
[3H]BrTuTP at specific activities of 105–106 cpm/lmol,
0.4 mM UTP and 5–15 U of E. coli RNA polymerase.
The reaction mixtures were incubated at 37 ꢁC and the
extent of synthesis determined by measuring the trichlo-
roacetic acid-precipitable radioactivity in 50 ll aliquots
on millipore filters. For the large scale synthesis of poly
r(BrTu-U) or poly r(BrTu) the reaction mixtures (5–
10 ml) were incubated at 37 ꢁC until the rate of polymer-
ization started to plateau. DNase I (10 lg/ml) was then
added and incubation continued for a further 30 min.
The reaction mixture was deproteinized with phenol.
The aqueous layer was extensively dialysed against
500–100 mM KCl, 10 mM Tris–HCl, pH 7.9, and was
used for subsequent experiments.
In conclusion, three aspects of BrTu action are impor-
tant: (1) BrTu reversibly inhibited cellular RNA synthe-
sis, but it did not inhibit either the picornavirus RNA
synthesis or multiplication. This distinction by BrTu be-
tween cellular DNA-directed and picornavirus RNA-di-
rected RNA synthesis should prove valuable for
studying the virus intracellular life cycle in the back-
ground of cellular RNA synthesis inhibition. (2) BrTu,
an adenosine kinase inhibitor, limited the amount of
highly cytotoxic Tu or DN that entered the cellular
nucleotide pools; the resulting intracellular concentra-
tion of toxic adenosine analogues was too low to inhibit
cell growth. (3) Since BrTu inhibited the synthesis of
high-molecular-weight cellular RNA, it protected the
cell against damage secondary to Tu or DN incorpora-
tion. This facet of 2-fold protection by BrTu probably
accounts for the excellent recovery of cells from the
combined treatment with BrTu and either of the two
cytotoxic analogues; the presence of BrTu simply nulli-
fied the cytotoxic action of the other nucleoside ana-
logues. Given the toxicity that accompanied the
incorporation of even small amounts of Tu into cellular
or viral nucleic acids, it was possible to achieve antiviral
selectivity10 by promoting Tu incorporation into picor-
navirus infected cells while preventing uptake into unin-
fected cells. Furthermore, it could also be possible to
exploit nucleoside transport and incorporation differ-
ences between different cell and tissue types for the ra-
tional application of these pharmacologically active
nucleoside analogues.
3.3. Enzymes
Escherichia coli RNA polymerase was prepared and as-
sayed according to Chamberlain and Berg.26 E. coli
alkaline phosphatase, pancreatic DNase, RNase, snake
venom phosphodiesterase and hexokinase were obtained
from Worthington Corp., Freehold, NJ, USA; myoki-
nase, phosphoenol pyruvate kinase were purchased from
Calbiochem, La Jolla, CA, USA. Rabbit liver adenosine
kinase was prepared and assayed as described by Lind-
berg et al.27 L-cells particulate and soluble fractions
were prepared in the following way: washed cells were
homogenized in a Dounce homogenizer in 10 mM Tris,
pH 7.2, containing 10 mM NaCl, 1.5 mM MgCl2 and
1 mM EDTA. The homogenate was centrifuged at
1000g for 10 min, the supernatant collected, and an ali-
quot saved as particulate enzyme preparation. The
remaining portion of the supernatant was then centri-
fuged at 100,000g for 1 h and the supernatant saved.
The conditions used for assaying adenosine kinase in
both preparations with various nucleosides as substrates
were as previously described27: the assay mixture (1 ml)
contained 20 mM KPO4, pH 5.8, 0.7 mM ATP,
0.25 mM phosphoenol pyruvate, 0.5 mM MgCl2,
3. Experimental
3.1. Chemicals
All pyrrolo-pyrimidine nucleosides were synthesized in
the laboratory of Dr. R.K. Robins, University of Utah,
Salt Lake City, and were generously provided by Drs.
H. Wood and R. Engle, National Cancer Institute,
Bethesda; tubercidin was obtained through the generos-
ity of Dr. G. Hitchings, the Welcome Research Labora-
tory, Tuckahoe, NY. The normal nucleosides and

B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
1491
50 mM KCl, 20 lg of pyruvate kinase, 7–70 lM of
radioactive nucleoside substrate and 25–100 lg of aden-
osine kinase preparation. The incubation was at 25 ꢁC
for various time intervals, the reaction was stopped by
adding 0.25 M perchloric acid, cooled in ice and electro-
phoresed on 3 MM Watman paper at 1500 V for 60–
90 min in 0.05 M citrate buffer, pH 3.5. The spots con-
taining nucleotides were cut out and the radioactivity
was measured in a liquid scintillation spectrometer.
red gently with a glass rod and spun down by a low
speed centrifugation (1000g, 3 min). Resulting pellet
was resuspended in HM containing 1% Triton X-100
and 0.5% deoxycholate, vortexed very shortly to get
homogeneous suspension and spun down as above. This
step was repeated once more with the pellet, which was
finally examined under phase microscope for the purity
of nuclei. All supernatants were combined and saved as
the cytoplasmic fraction. The clean nuclei were lysed in
high salt buffer containing 10 mM Tris, pH 7.4, 500 mM
NaCl, 10 mM MgCl2 (2 ml for 30 · 106 cells), vortexed
shortly and treated with DNase I (50 lg/ml), which
caused the clearing of the mixture. Following centrifuga-
tion at 15,000g for 5 min the pellet was saved as nucleo-
lar fraction, and the supernatant represented
nucleoplasmic fraction.
3.4. Cells and viruses
Mouse fibroblasts (strain L2) were maintained and
propagated as monolayer cultures in minimal essential
medium (MEM) supplemented with 5% foetal bovine
serum (Gibco/BRL Life Technologies, Gaithersburg,
MD). Vaccinia virus and mengovirus were grown and
assayed as previously reported.2,10
RNA was extracted from nucleolar, nucleoplasmic and
cytoplasmic fraction with phenol–sodium dodecyl sul-
fate as follows: nucleoplasmic fraction was precipitated
with 2.2 vol of ethanol in cold and spun down at
10,000g for 10 min. The pellet was mixed with 2 ml of
TNE–SDS buffer (10 mM Tris, pH 7.4, 100 mM NaCl,
1 mM EDTA, 0.5% SDS). The clean nucleoli were
resuspended in 2 ml of the same buffer, and the cyto-
plasmic fraction was made 0.5% in SDS. From that
point on, all three fractions were treated equally to ex-
tract RNA with phenol–SDS and were analysed by elec-
trophoresis on 2.5% acrylamide gels as previously
described.12
3.5. Macromolecule synthesis
The incorporation of radioactive precursors into cellular
macromolecules was followed by the procedures de-
scribed previously.2,12 In brief, cells from monolayer cul-
tures exposed to the radioactive precursors for varying
periods were harvested and washed with ice-cold saline.
They were then extracted with 0.25 M perchloric acid
for 45 min with occasional shaking. After centrifugation
the acid-soluble material was adsorbed on Norit for 1 h
at 4 ꢁC, the Norit-adsorbable nucleotides were then
eluted in 50% ethanol containing 2% ammonia, and
the aliquots were measured for OD260nm as well as
radioactivity in scintillation fluid. The acid-insoluble
fraction was dissolved and incubated in KOH (0.5 M,
37 ꢁC, 18 h); the precipitate which formed on acidificat-
ion of the alkaline solution contained the cellular DNA,
whereas the material which remained soluble repre-
sented the degraded cellular RNA. The DNA fraction,
which was degraded by perchloric acid (0.5 M, 90 ꢁC,
30 min), and the soluble RNA fraction were measured
for radioactivity in scintillation fluid.
Acknowledgment
This work was supported in part by Ministry of Science,
Education and Sports of the Republic of Croatia, Grant
No. 0098078.
References and notes
1. Acs, G.; Reich, E.; Mori, M. Proc. Natl. Acad. Sci. U.S.A.
1964, 52, 493.
2. Brdar, B.; Reich, E. J. Biol. Chem. 1972, 247, 725.
3. Tavitian, A.; Uretsky, S.; Acs, G. Biochim. Biophys. Acta
1968, 157, 33.
4. Suhadolnik, R. J. In Nucleosides as Biological Probes;
Wiley-Interscience: New York, 1979; pp 158–168.
5. Ward, D. C.; Reich, E.; Stryer, L. J. Biol. Chem. 1969,
244, 1228.
6. Ward, D. C.; Reich, E. J. Biol. Chem. 1972, 247, 705.
7. Pope, L. H.; Shotton, M. W.; Forsyth, T.; Hughes, D. J.;
Denny, R. C.; Fuller, W. Biophys. Chem. 1998, 70, 161.
8. Langen, P. R.; Waldrop, A. A.; Ward, D. C. Proc. Natl.
Acad. Sci. U.S.A. 1981, 78, 6633.
9. Davis, L. P.; Jamieson, D. D.; Baird-Lambert, J.;
Kazlavskas, R. Biochem. Pharmacol. 1984, 33, 347.
10. Brdar, B.; Reich, E. J. Virol. 1999, 73, 6444.
11. Brdar, B.; Reich, E. Period. Biol. 1976, 78, 51.
12. Brdar, B.; Rifkin, D. B.; Reich, E. J. Biol. Chem. 1973,
248, 2397.
13. Henderson, J. F.; Paterson, A. R. P.; Caldwell, I. C.; Paul,
B.; Chan, M. C.; Lau, K. F. Cancer Chem. Rep. 1972, 3, 71.
14. Parkinson, F. E.; Geiger, J. D. J. Pharmacol. Exp. Ther.
1996, 277, 1397.
The procedure used for monitoring viral RNA of
mengovirus-infected L-cells was previously reported.2,10
In brief, monolayers of L-cells were incubated with acti-
nomycin D (2 lg/ml) for 20–30 min. After this, the med-
ium was removed and the cells were infected with
mengovirus at a multiplicity of infection of 10–100. Fol-
lowing the period of virus adsorption (60 min, 37 ꢁC) the
cells were incubated in fresh medium containing actino-
mycin D and radioactive RNA precursor. At the end of
periods of incubation the cells were washed, and the
radioactivity in RNA alkali labile fraction or trichloro-
acetic acid-precipitable material was determined in a
scintillation counter.
3.6. Cell fractionation and RNA extraction
Nucleolar, nucleoplasmic and cytoplasmic fractions
were isolated according to a modified procedure re-
ported earlier.12 In brief, washed cells were resuspended
in homogenizing media (HM) (10 mM Tris, pH 7.2,
150 mM NaCl, 2 mM MgCl2, 0.05% Triton X-100), stir-

1492
B. Brdar, E. Reich / Bioorg. Med. Chem. 16 (2008) 1481–1492
15. Cass, C. E.; Young, J. D.; Baldwin, S. A.; Cabrita,
M. A.; Graham, K. A.; Griffiths, M.; Jennings, L. L.;
Mackey, J. R.; Ng, A. M.; Ritzel, M. W.; Wickers,
M. F.; Yao, S. Y. Pharmacol. Biotechnol. 1999, 12,
313.
20. Brdar, B.; Rubelj, I.; Reich, E. Period. Biol. 1999, 101,
131.
21. Ossowski, L.; Reich, E. Virology 1972, 50, 630.
22. Kapuler, A. M.; Ward, D. C.; Mendelsohn, N.; Klett, R.;
Acs, G. Virology 1969, 37, 701.
16. Van Rompay, A. R.; Johansson, M.; Karlsson, A.
Pharmacol. Ther. 2003, 100, 119.
23. Kapuler, A. M.; Spiegelman, S. Proc. Natl. Acad. Sci.
U.S.A. 1970, 66, 539.
17. Rogler-Brown, T.; Parks, R. E., Jr. Biochem. Pharmacol.
1980, 29, 2491.
24. Darlix, J. L.; Fromageot, P.; Reich, E. Biochemistry 1971,
10, 1525.
18. Kong, W.; Engel, K.; Wang, J. J. Curr. Drug Metab. 2004,
5, 63.
19. Sinclair, C. J. D.; Powell, A. E.; Xiong, W.; LaRiviere, C.
G.; Baldwin, S. A.; Cass, C. E.; Young, J. D.; Parkinson,
F. E. Br. J. Pharmacol. 2001, 134, 1037.
25. Plagemann, P. G. J. Cell. Physiol. 1971, 77, 241.
26. Chamberlin, M.; Berg, P. Proc. Natl. Acad. Sci. U.S.A.
1962, 48, 81.
27. Lindberg, B.; Klenow, H.; Hansen, K. J. Biol. Chem. 1967,
242, 350.