Download Inhibition of the Particle-associated RNA

J. gen. Virol. 0974), 23, 59-7I
59
Printed in Great Britabl
Inhibition of the Particle-associated
RNA-dependent R N A Polymerase Activity of Influenza
Viruses by Chelating Agents
By J. S. O X F O R D * AND D. D. P E R R I N
Department of Microbiology and Department of Medical Chemistry,
The John Curtin School of Medical Research, The Australian National University,
Canberra, Australia
(Accepted 16 November I973)
SUMMARY
Approximately o'o5 mM concentrations of the chelating agents bathocuproine
and bathophenanthroline disulphonic acid disodium salts, which form very stable
complexes with 'soft' heavy-metal ions such as zinc, inhibited in vitro the RNAdependent RNA polymerase activity associated with influenza B/LEE/4o and
A]RI-5 + (H2N2) viruses in the presence of a large molar excess of Mn(II) and
Mg(II). Certain heterocyclic thiosemicarbazones also inhibited influenza RNA
polymerase activity, although complete inhibition was not detected even with high
concentrations (o'I raM) of the compounds. Analogues of the active heterocyclic
thiosemicarbazones, such as isatin 3-semicarbazone, which have reduced chelating
activities for zinc, also had reduced inhibitory effects on influenza virus associated
RNA polymerase activity. However, inhibition of influenza RNA polymerase
activity by o.I mM-bathocuproine or o.t mM-isatin 3-thiosemicarbazone was not
reversed by the addition of molar equivalent or excess concentrations of zinc.
Relatively high concentrations of the chelating agents had no detectable effect on
the haemagglutinin, neuraminidase or infectivity titres of influenza B/LEE[4o
virus. The chelating agents may inhibit influenza virus-associated RNA polymerase
activity by the formation of a ternary complex of enzyme-metal-ligand.
INTRODUCTION
Influenza A and B viruses have RNA-dependent RNA polymerase activity, associated with
virus cores (Chow & Simpson, I97I ; Skehel, 1971). The polymerase enzyme can transcribe,
at least in vitro, a portion of the influenza virus genome, suggesting that the enzyme has an
important early function in cell infection, such as the synthesis of influenza messenger RNA
species (Bishop et aL 1972). The RNA-dependent RNA polymerase enzyme might therefore
provide a very useful and specific site of action for the development of chemoprophylactic
agents against viruses containing RNA (Ho et al. 1967; Ho & Walters, 1971).
A previous study of the inhibition of influenza A and B virus RNA polymerase by selenocystamine dihydrochloride indicated that the compound acted either by oxidizing -SH
groups of the RNA polymerase enzyme polypeptide(s), or by chelating an essential heavy
metal of a metalloenzyme or metal activated polymerase enzyme (Oxford, I973a ).
Other studies have demonstrated that some cell-associated DNA and RNA polymerases
* Present address : Division of Virology, National Institute for Medical Research, Mill Hill, London,
N.W.7.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
60
J.S.
O X F O R D AND D. D. P E R R I N
are zinc meta!loenzymes. Thus, purified preparations of DNA polymerase from Escherichia
coli and from Strongylocentrotus franciscanus blastulae have been shown by atomic absorption spectroscopy to contain 2 and 4 g atoms of zinc per tool of enzyme, respectively (Slater,
Mildvan & Loeb, I971), and enzyme activity is inhibited by the chelating agent orthophenanthroline. Further, zinc is present in purified preparations of RNA polymerase of
E. coli (Scrutton, Wu & Goldthwait, I97I) and may participate in the initiation of RNA
synthesis.
If the RNA-dependent RNA polymerase of influenza virus is a metalloenzyme, activated
by metal ions other than magnesium or manganese, some indication of the nature of the
metal ion involved could be expected from a study of the inhibitory effects of a range of
chelating agents or ligands. The ligands selected for the present study were able to chelate
different metals, depending on whether the donor atoms were nitrogen and sulphur or
nitrogen and oxygen (Perrin, 197o). Further, in view of the earlier reports on the use of
isatin 3-thiosemicarbazone and its derivatives as antiviral agents (Sadler, 5963; Bauer, 1965)
and the activity of these compounds as chelators of metal ions, particularly zinc and copper
(Gingras, Suprunchuk & Bayley, I962), we have examined the effects of certain heterocyclic
thiosemicarbazones on influenza virus RNA polymerase activity.
METHODS
Viruses. Influenza strains A/RI-5 + (H2N2) and B/LEE[4o were supplied by Dr W. G.
Laver, Department of Microbiology, The John Curtin School of Medical Research,
Canberra, Australia.
Virus purification. Influenza viruses were grown in Io-day-old embryonated hens' eggs
and purified as described previously (Laver, I969), with the addition of an equilibrium sedimentation of the virus in potassium tartrate (Oxford, I973b ). Standard preparations of
influenza B/LEE]4o had asp. act. of 3"6 × IO~ H.A.U./mg protein with an overall purification factor of 4oo.
RNA polymerase and RNA polymerase inhibition tests. The reaction mixture (IOO #1)
contained 5o mi-tris-HC1 buffer, pH 8.o, 5 mi-MgCl~, 2 mM-MnC12, o'5 mM-GTP, ATP,
CTP, o.oo5 mM-[3H]-UTP (approx. 7"5 #Ci), o-2 ~ (v/v) Nonidet P4o and 50/*g of purified
influenza virus protein. The polymerase reaction was initiated by the addition of virus and
incubated at 37 °C for 3o min. For most virus preparations incorporation of [3H]-UMP
into acid-precipitable product increased linearly during this period (J. S. Oxford, unpublished
data). At zero time, and after 30 min, duplicate 2o/*l samples were pipetted on to Whatman
grade 3 MM filter papers which were air-dried at 5o °C for 15 min and then washed
successively in 6 ~ (v/v) TCA containing o.I i-pyrophosphate, 6 ~ (v/v)TCA, ethanol and
ether. The dried discs were counted for radioactivity in a Packard liquid scintillation
spectrometer using IO ml of a toluene solution containing o'5 ~ PPO and 0"o3 ~ POPOP.
The RNA-dependent RNA polymerase activity associated with purified influenza
particles in the present study was found to be dependent on the presence of Mn 2+ as
described previously (Oxford, I973a ). Both Mg 2+ and Mn ~+ were added to the reaction
mixture as described by Chow & Simpson (i97 0. Skehel (1971) reported that either Mn 2+
or Mg 2+ stimulated the reaction depending on the temperature of incubation, but this effect
was not investigated in the present study. Approx. 640ooo d]min of [~H]-UMP]mg influenza
virus protein were incorporated into acid-precipitable product during 30 rain at 37 °C, using
influenza B]LEE/4o virus. Since throughout the study different virus preparations were
used, a single inhibitory compound (selenocystamine dihydrochloride) was included in
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
Inhibitors o f influenza R N A polymerase
6I
every test as a standard reference compound. For accurate quantitative comparisons,
various compounds were tested in parallel in the same experiment using a single virus preparation. Unless otherwise stated, inhibitory compounds at a range of different concentrations from o-I to o.ooi mM were added last to the RNA polymerase reaction mixture at 4 °C,
immediately before warming to 37 °C. The concentration of compound inhibiting incorporation of [3H]-UMP into acid-precipitable product by 5o ~ was derived from the dose response curve obtained for each inhibitor.
Chemicals. The disodium disulphonate salts of bathocuproine (4,7-diphenyl-2,9-dimethylI,Io-phenanthroline) and bathophenanthroline (2,9-dimethyl-I,to-phenanthroline), and
diethylenetriamine were obtained from Fluka AG (Switzerland). 4-(z-Pyridylazo) resorcinol disodium salt (PAR) was obtained from Eastman Organic Chemicals, Rochester,
New York. Other chemicals were obtained from British Drug Houses, Poole, Dorset.
The thiosemicarbazones were prepared by direct condensation of aqueous or aqueous
ethanolic solutions of thiosemicarbazide and the aldehyde or ketone by standard methods
(Mann & Saunders, I96O), followed by recrystallization from aqueous ethanol and drying
at IiO °C. Similarly, for the semicarbazones, semicarbazide hydrochloride in an acetic
acid-sodium acetate buffer was used and the amidinohydrazones were made by condensation
of the appropriate ketone with aminoguanidinium bicarbonate (Mann & Saunders, I96O).
Isatin 3-selenosemicarbazone was synthesized in a manner analogous to thiosemicarbazones
by Dr W. Pendergast, Department of Medical Chemistry, The John Curtin School of
Medical Research, Canberra. The thiosemicarbazones were relatively insoluble in aqueous
solutions and were therefore dissolved at I mg/ml in 50 ~ (v/v) dimethyl sulphoxide
(DMSO) and deionized water. All solutions of inhibitory compounds were prepared
freshly for testing. Separate tests established that the final concentrations of DMSO in the
reaction mixture had no effect on influenza RNA polymerase activity.
Unlabelled deoxy- and ribonucleoside triphosphates and methyl-[aH]-TTP (I 6.8 Ci/mmol)
were purchased from Schwarz-Mann, Orangeburg, New York, and [3H]-UTP (I 5"9 Ci/mmol)
from New England Nuclear, Boston, Mass.
RNA polymerase from Escherichia coli K-I2 and DNA polymerase from Micrococcus
lysodeikticus were purchased from Miles Laboratories Inc., Elkhart, Indiana. Calf thymus
DNA was purchased from Worthington Biochemical Corporation.
RESULTS
A wide range of heavy-metal chelating agents were tested for any inhibitory effect on the
RNA-dependent RNA polymerase activity of influenza B/LEE/4o virus. Metal chelating
agents were selected particularly which would have little chelating activity for manganese
and magnesium ions but which would form very stable complexes with heavy metals such
as zinc, copper and iron (SiUen & Martell, 197 0. These latter three metals are the ones
commonly found in metalloenzymes and heavy metal-activated enzymes (Williams, 197I).
Although some of the ligands tested would chelate Mn(II) and Mg(II) under the test conditions of influenza RNA polymerase enzyme activity, a molar excess of both Mn(II) and
Mg(II) were present compared to the concentrations of chelating agents. Thus inhibition of
enzyme activity could not be attributed to removal of the latter two cations, which have
been shown to be necessary for the demonstration of influenza RNA polymerase activity
in vitro (Chow & Simpson, 1971). Influenza B]LEE/4o virus was used for most experiments
because the virus multiplied to high titres in eggs and could be purified with relative ease
(Oxford, I973b ).
5
VXR 23
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
62
J.S.
OXFORD AND D. D. PERRIN
Table ~. Effect of chelating agents on RNA-dependent RNA polymerase
activity of influenza B[LEE[4o virus
Chelating agent
L With donor N atoms
Bathophenanthroline disulphonic
acid disodium salt
Bathocuproine disulphonic acid
disodium salt
Triethylenetetramine
Diethylenetriamine
2 : 2'-dipyridyl
2. With donor S atoms
Cystamine
Diphenylthiocarbazone (dithizcne)
Penicillamine
fl-mercaptoethanol
2,3-dimercapto-I-propanol (BAL)
3. With mixed donor atoms
Solochrome black
Eriochrome blue
Zincone
PAR
8-hydroxyquinoline
Concentration (raM) inhibiting
B/LEE/4o virus RNA polymerase
activity by 50 ~*
0"05_+o.o2
o-o8_+o-o2t
0"93• o"I
4"66_+2.18
6"4_+o"lO
o'3
o'4
6'7
ioo.o
2oo.o
o. 13+-o'o3
o. 15-+0"05
o. I9 _+0"07
0.26 + 0.08
I "30
* The RNA-dependent RNA polymerase enzyme associated with the virus preparations incorporated
approx. 64oo0o d/rain [SH]-UMPinto acid-insoluble product/mg virus protein/3o min at 37 °C.
? Standard errors in this and subsequent tables are given for five or more determinations. Other figures
represent mean values of two or three determinations.
RNA-dependent R N A polymerase activity of influenza B/LEE/4o virus particles was
inhibited by a number of chelating agents having donor N atoms available for chelation of
metal ions (Table I). Particularly effective inhibitors of enzymic activity were bathocuproine
and bathophenanthroline disulphonic acid disodium salts, which inhibited influenza
B/LEE/4o R N A polymerase activity by 5o ~ , at concentrations of o'o8 mM and o'o5 re_u,
respectively. Higher concentrations of these two compounds inhibited R N A polymerase
activity completely, and a dose response curve for bathocuproine disulphonic acid disodium
salt is shown in Fig. I.
Chelating agents containing sulphur were less inhibitory than bathophenanthroline
(Table I). The most active compound in this group was cystamine, which inhibited influenza
B/LEE/4o virus R N A polymerase activity by 5o ~ at o'3 mM.
A series of eriochrome dyes and compounds chelating metals through sulphur and
nitrogen or nitrogen and oxygen atoms were also tested for inhibitory effect on influenza
B/LEE R N A polymerase activity (Table I). Of these, the most active inhibitors were the
eriochrome dyes.
Thus, the most active inhibitors of influenza B/LEE/4o RNA-dependent R N A polymerase activity selected from the above three groups of metal chelating agents were
bathocuproine and bathophenanthroline. These two compounds are particularly effective
chelators of zinc and Cu(I) metal ions (Sillen & Martell, I964). Isatin 3-thiosemicarbazone
and its derivatives are also active chelating agents of zinc and copper (Gingras et al. I962),
therefore representative heterocyclic thiosemicarbazones were tested for any inhibitory
effect on influenza R N A polymerase activity.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
Inhibitors of influenza RNA polymerase
I
I
I
.>_
I
:/i
N 80
<
z
63
60
ts~
..d
g
40
#
20
I
I
0.04
I
O. 16
I
0.64
2.56
Conc. of inhibitor (raM)
Fig. I. Dose-response curve for inhibition of influenza B/LEE/4o virus associated RNA polymerase
activity by the disodium disulphonate salt of bathocuproine.
Effect of heteroeyclic thiosemicarbazones and analogues on RNAdependent RNA polymerase activity of influenza B/LEE]4o virus
Table 2.
Compound
Isatin 3-thiosemicarbazone
Isatin 3-selenosemicarbazone
Methyl isatin 3-thiosemicarbazone
Isatin 4"-metbylthiosemicarbazone
Isatin 4'-ethylthiosemicarbazone
Isatin 3-semicarbazone
Isatin 3-amidinohydrazone
z-acetylpyridinethiosemicarbazone
3-acetylpyridine thiosemicarbazone
4-acetylpyridine thiosemicarbazone
z-acetylpyridine semiearbazone
Concentration of compound (mM)
inhibiting B/LEE RNA polymerase
activity by 5o ~
0"oo4
0'03 + 0.008
0"02 ~
O'IO
0"I0
0-04
>0'5
>o'5
0 " 0 0 2 -t- 0 ' 0 0 I
I'0
I'0
I'O
Inhibitory effect of heterocyclic thiosemicarbazones
I s a t i n 3-thiosemicarbazone, 2-acetylpyridine t h i o s e m i c a r b a z o n e a n d isatin 3-selenosemic a r b a z o n e were effective inhibitors o f R N A p o l y m e r a s e activity o f influenza B / L E E ] 4 o
virus (Table 2) a n d were selected as representative c o m p o u n d s for further studies. The d o s e response curves for the three c o m p o u n d s are shown in Fig. 2 a n d i n d i c a t e d t h a t influenza
5-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
J. S. OXFORD AND D. D. PERRIN
64
I
100
;~
!
!
I
I
80
_
z¢.
/.
I
60
<
z
,-1
.d
~- 40
¢-.
.E
20
/
S"
I
0"003
I
0"012
0.048
Conc. of inhibitor (raM)
0' 192
Fig. 2. Dose-response curves of the inhibitory effects of heterocyclic thiosemicarbazones on the
RNA-dependent RNA polymerase activity of influenza B/LEE/4o virus : , - - T , isatin 3-selenosemicarbazone; D--D, isatin 3-thiosemicarbazone; i - - i , 2-acetylpyridine thiosemicarbazone.
H
I
s
NH2
Fig. 3, Formation ofa terdentate complexby chelation of metal ion (M)
with isatin 3-thiosemicarbazone.
B/LEE/4o virus particle associated R N A polymerase activity was not suppressed completely
by even high concentrations of 2-acetylpyridine thiosemicarbazone or by isatin 3-thiosemicarbazone. Thus, R N A polymerase enzyme activity was inhibited maximally (80 %) by
o"o9 mM-isatin 3-thiosemicarbazone, a concentration approaching the maximum solubility
of the compound at 37 °C (o'I9 mM). In contrast, the dose-response curve for isatin 3selenosemicarbazone indicated that influenza B/LEE/4o virus R N A polymerase activity
could be inhibited completely by o'o9 mM of this compound (Fig. 2). Isatin 3-thiosemicarbazone exerts its maximum metal binding effect when it forms a terdentate chelate with a
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
Inhibitors of influenza R N A polymerase
65
Table 3. Lack of virucidal effect of chelating agents on influenza B/LEE]4o virus
Compound
Control
Bathophenanthroline disulphonic acid
disodium salt
Isatin 3-thiosemicarbazone
Isatin 3-selenosemicarbazone
2-acetylpyridine thiosemicarbazone
Selenocystamine dihydrochloride
Concentration of
compound (mM)
Infectivity titre, log (EIDso/ml),
after incubation for 30 min at 37 °C
0
3"76+o'I3
2.o
o.2o
0.20
o.20
o-3o
3"8o+oq5
3"72 + o'19
3"69 _+o-o8
4'7
1.65+0.I 5
metal ion by loss of a proton from the thioenolic form and by co-ordination of the metal
through the sulphur, nitrogen and oxygen atoms as shown in Fig. 3. Bidentate chelation
involving only two of these atoms would be expected to lead to appreciably less stable
complex formation. For copper(I) and zinc(II) the expected stability constant sequence for
related complexes is Se > S /> o ~ N. It would be expected therefore that isatin 3-selenosemicarbazone should be a particularly effective chelator of these two metals; this compound,
as described above, unlike other heterocyclic thiosemicarbazones, was able to cause
complete inhibition of influenza B/LEE RNA polymerase activity. Conversely, isatin 3semicarbazone and isatin 3-amidinohydrazone in which the sulphur atom in the thiosemicarbazone portion of the molecule is replaced by oxygen and nitrogen respectively should
have markedly reduced chelating activity for Cu(I) and zinc(II) (Perrin, ~97o). These
analogues were synthesized and found to have significantly diminished inhibitory effect on
influenza B/LEE/4o virus associated RNA polymerase activity (Table 2).
The chelating agent 2-acetylpyridine thiosemicarbazone has the correct stereochemistry
for terdentate complex formation with zinc or copper(I). The compound inhibited B/LEE/4o
RNA polymerase activity by 50 ~ at a concentration of o.oo2 mN (Table z). This enzyme
inhibitory activity was much reduced in the 3- and 4-acetylpyridine analogues (Table 2) in
which metal chelation involving the ring nitrogen is no longer possible because of stereochemical effects. It was also reduced in 2-acetylpyridine semicarbazone, which would have
diminished chelating properties because of the replacement of sulphur in the thiosemicarbazone portion of the molecule. A correlation was therefore established between the chelating
properties of heterocyclic thiosemicarbazones for heavy-metal ions and the inhibitory effects
on influenza RNA polymerase activity.
Mode of action studies with selected metal chelating agents
The disodium disulphonate salts of bathophenanthroline and bathocuprione had no
detectable inhibitory effect in four separate experiments on the haemagglutinin titre, neuraminidase activity (Warren, 1959) or infectivity titre of influenza B/LEE/4o estimated by the
allantois on shell technique (Fazekas de St Groth & White, 1958). Relatively high concentrations of the compounds (2 mM) were incubated for 30 rain at 37 °C with influenza virus in
o'o5 M-tris HC1 buffer at pH 8"o and the mixtures then rapidly diluted and titrated for
residual biological activities. Selenocystamine dihydrochloride (o'o3 mM) when incubated
as a control reduced the infectivity titre of influenza B/LEE/4o virus by 2.I log10units (Table
3) confirming previous results with this compound (Oxford, I973a ). Relatively high
concentrations (o'z mM) of isatin 3-thiosemicarbazone,/-acetylpyridine thiosemicarbazone
and isatin 3-selenosemicarbazone had no detectable effect on virus infectivity, or on the
haemagglutinin or neuraminidase titres of influenza B/LEE/4o virus. Preliminary experi-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
66
J . S. O X F O R D
AND
D.
D.
PERRIN
ments have indicated that extensive dialysis of isatin 3-semicarbazone and influenza virus
mixtures results in the restoration of virus-associated R N A polymerase activity and this
may explain the absence ofvirucidal effect of these compounds (J. S. Oxford & D. D. Perrin,
unpublished observations).
Addition of excess reagents in the RNA polymerase reaction mixture, including 4 mMMn 2+, 6 mM-Mg 2+ and I'5 mM-ATP, CTP, GTP and o'oi5 mM-[3H]-UTP failed to prevent
or reverse the inhibitory effect of o.I mM-bathocuproine or o.o8 mM-isatin 3-thiosemicarbazone on the RNA-dependent polymerase activity of B/LEE]4o virus suggesting that the
compounds were not acting simply by removing essential added reagents from the reaction
mixture.
The disodium disulphonate salts of bathocuproine or bathophenanthroline were added at
varying times (final concentrations of o.~ raM) to the complete reaction mixture for the
detection of RNA polymerase enzyme activity of influenza B/LEE/4o virus. Enzyme
activity, as detected by incorporation of [~H]-UMP into acid-precipitable product, was
inhibited immediately if the compounds were added at 5 or Io min after the reaction
mixture had been incubated at 37 °C. At these times a rapid and continuing increase in
enzyme activity was detected in control preparations with no inhibitor added. Similarly,
isatin 3-thiosemicarbazone (o.o8 mM) added 5 min after the B/LEE/4o virus RNA polymerase reaction had been initiated at 37 °C, could still inhibit incorporation of [~H]-UMP
into acid-precipitable product. These experiments indicated that pre-incubation of the
chelating agents with virus was not necessary in order for inhibitory effects to be detected.
No further increase in inhibitory activity of the chelating agent was detected by incubating
virus with bathocuproine disulphonic acid disodium salt for I or 2 h at 4 °C before the
reaction mixture was incubated at 37 °C to initiate RNA polymerase enzyme activity.
Attempts were made to reverse or prevent the inhibitory effects of o'5 mM-bathocuproine
or o-~ mM-heterocyclic thiosemicarbazones on the RNA polymerase activity of influenza
B/LEE/4o virus by the addition of Io mM-fl-mercaptoethanol or 3 mM-dithiothreitol to the
reaction mixture. Such experiments might indicate a role of SH groups in RNA polymerase
activity and inhibition. In control reaction mixtures both dithiothreitol and fl-mercaptoethanol at these concentrations stimulated incorporation of [3H]-UMP into acid-precipitable
product by influenza B/LEE/4o virus by approx. 5o ~. The inhibitory effects of isatin 3thiosemicarbazone, isatin 3-selenosemicarbazone and 2-acetylpyridine thiosemicarbazone
were reversed completely by the addition of a two- to fivefold molar excess of fl-mercaptoethanol to the reaction mixture. A similar reversal effect by fl-mercaptoethanol was noted
previously with selenocystamine (Oxford, I973a), although in the latter case fl-mercaptoethanol possibly acted by a reductive scission of diselenide to selenol. The inhibitory effect
of bathocuproine on B/LEE/4o RNA polymerase activity was not reversed by the addition
of up to a fivefold molar excess of fl-mercaptoethanol or dithiothreitol.
Spectrum of inhibitory activity of the chelating agents for virus and
bacterial polymerases
We wished to determine whether the chelating agents had a selective effect against
influenza virus associated RNA polymerase, which would be the optimum property of any
effective chemoprophylactic agent against R N A viruses, or whether the compounds had a
wider range of inhibitory activity for cellular RNA or D N A polymerase enzymes. Selected
compounds were tested for inhibitory effect on the DNA-dependent R N A polymerase from
Eseherichia coIi KI2 and the DNA-dependent D N A polymerase from Micrococcus lysodeik-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
Inhibitors o f influenza R N A polymerase
67
Table 4. Effect of chelating agents on virus and bacterial RNA and DNA polymerases
Conc. of compound (raM) to inhibit incorporation of
13H]-UMP or [3H]-TMP by 5o
Micrococcus
A/RI-5 + (H2N2)
Escherichiacoli
lysodeikticusC
RNA polymerase* RNA polymeraset DNA polymerase
Compound
Bathocuproine disulphonic acid disodium salt
Eriochrome blue
Zincone
Isatin 3-thiosemicarbazone
2-acetylpyridine thiosemicarbazone
o.oz
o.ot
o. Io
0.06
o'oo3
o.o8
o.o8
o"~o
o'3o
Not tested
Not tested
> 0"20
> 0"20
> o.2o
> o.zo
* Approx. 140000 d/min incorporated per mg virus protein/3o rain at 37 °C.
t Standard reaction mixture used with addition of 2/zgofE. coliKi2polymeraseproteinand 2"5#gofcalf
thymus DNA. Approx. zo x Io 4 d/min incorporated per mg protein/3o min at 37 °C.
Reaction mixture (IOO/zl) contained: 5o mM-tris-HCl buffer, pH 8"o, 5 mM-MgCI~, o'5 mM-dGTP,
dATP, dCTP and o'oo5 mM-methyl-[ZH]-TTP,6/~g of M. lysodeikticus polymerase protein and 2"5/zgof calf
thymus DNA. Approx. 8z x Io 4 d/rain incorporated per mg protein/3o min at 37 °C.
ticus, in the absence of a source of purified DNA-dependent R N A polymerase from
mammalian cells. The inhibitors were added to the reaction mixture at time zero, followed
immediately by influenza virus or bacterial polymerases.
N o inhibitory activity for bacterial DNA-dependent D N A polymerase or DNA-dependent
R N A polymerase was detected when isatin 3-thiosemicarbazone, isatin 3-selenosemicarbazone or 2-acetylpyridine thiosemicarbazone were added to the reaction mixtures at concentrations of o-2o mM (Table 4). However, eriochrome blue and bathocuproine inhibited the
DNA-dependent R N A polymerase of Escherichia coli K I 2 at concentrations of o-o8 mM approx, only eight- and fourfold the concentrations required by these compounds to inhibit
RNA-dependent R N A polymerase of influenza A/RI-5 + (H2N2) virus.
Isatin 3-thiosemicarbazone and z-acetylpyridine thiosemicarbazone have inhibitory
activity against at least one influenza A virus R N A polymerase, since the particle-associated
R N A polymerase of influenza A/RI-5 + (HzN2) was inhibited by 5o ~ by the addition of
o.o6 mM and o'oo3 mM, respectively, of the compounds. Bathocuproine disulphonate
disodium salt at a concentration of o.o2 mM also inhibited the R N A polymerase activity
associated with influenza A/RI-5 + (HzN2) virus (Table 4).
Inhibitory activity of ligand-metat complexes
In an attempt to determine if the chelating agents inhibited the influenza virus R N A
polymerase enzyme activity by removing an essential metal ion from the enzyme polypeptide(s) or alternatively by masking a metal in situ by the formation of an enzyme-metalinhibitor complex, experiments were carried out by adding preformed ligand-metal
complexes to the reaction mixture. I f any essential metal was actually removed from the
polymerase enzyme by the chelating agent, addition of preformed ligand-metal complex
would be expected to have little inhibitory effect on the polymerase enzyme activity because
the chelating agent would be already saturated with metal. However, ~ : I and 2: ~ complexes
of bathocuproine and Cu(I) prepared as described by Hall, Marchant & Plowman (I962)
inhibited the R N A polymerase activity of influenza B/LEE/4o virus by 5o ~ at o.oz raM,
and hence were rather more active than bathocuproine itself. The complex bis(z,9-dimethylI, I o-phenanthroline) Cu(I) sulphate, the ~ : I bathocuproine-Cu(I) complex (prepared in the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
68
J.
S. O X F O R D
AND
D.
D.
PERRIN
Table 5. Effect of heavy-metal ions on RNA-dependent RIgA polymerase
activity of influenza B[LEE/4o virus
Metal
Cu(II)
Zn(II)
Fe(III)
Co(II)
Hg(II)
Conc. (raM)to inhibit incorporation
of pH]-UMP by 50
0"08-t-o'oI
0-06+ o.o2
0"83+ o. IO
0'85 + 0'58
I "25+ o'75
presence of to mM-hydroxylamine to ensure that copper was present in the complex as
Cu(I)), and the ~ : I bathocuproine-zinc complex, also inhibited influenza B/LEE/4o RNA
polymerase activity by 5o ~o at concentrations of 0.02 mM, O'm and o'oI mN, respectively.
Similar results were obtained in the present study with ~ : t complexes of copper(I), zinc,
silver and cadmium and 2-acetylpyridine thiosemicarbazone, synthesized as described by
Gingras et al. 0962). These complexes inhibited influenza B/LEE/4o R N A polymerase
activity by 5o ~ at concentrations of o'ooi mN and furthermore, in contrast to the parent
compound, 2-acetylpyridine thiosemicarbazone, higher concentrations of the complexes
(approx. o.o 5 mM) caused complete inhibition of influenza RNA polymerase activity.
Therefore, if the ligands inhibit influenza R N A polymerase enzyme activity by chelating an
essential metal ion such as zinc or copper, it would appear that the metal ion remains
attached to the enzyme polypeptide(s) and a ternary complex of enzyme, metal and ligand
is formed in situ. This complex might be expected to cause inhibition of enzyme function.
In view of the inhibitory activity of the bathocuproine-metal complexes, the possible
inhibitory effect of a number of metal ions on influenza B/LEE/4o RNA polymerase activity
was investigated. The results (Table 5) indicated that copper(H) and zinc(II) were particularly
effective inhibitors of influenza virus associated R N A polymerase enzyme activity, causing
5o ~o enzyme inhibition at concentrations of o.o8 and o.o6 raM, respectively.
The inhibition of certain zinc metalloenzymes by chelating agents can be reversed by the
addition of zinc (Vallee, t96o). Therefore, varying concentrations of either zinc or copper
were added to the influenza B/LEE/4o RNA polymerase reaction mixture containing o.I
mM-bathocuproine disulphonic acid disodium salt or o.I mN-isatin 3-thiosemicarbazone, in
attempts to reverse the inhibitory effects of these compounds. In four experiments, no
reversal of the inhibitory effects of the chelating agents was obtained when concentrations
of zinc between o.o2 and 2 mM were added both before and after the addition of the chelating
agents to the virus containing RNA polymerase reaction mixture.
DISCUSSION
The comparable electronegativities of sulphur and selenium, together with the known
chelating abilities of 2,2'-diaminodiethyl sulphide (Gonick, I95I ; Gonick, Fernelius &
Douglas, I954), 3,3'-dithiodipropanoic acid and cystamine (Hawkins & Perrin, t963),
which are structurally related to selenocystamine, makes it possible that the inhibition of
influenza virus RNA polymerase by selenocystamine (Oxford, t973a ) is due to the chelation
of a heavy-metal ion essential for enzyme activity. Because sulphur and selenium form their
most stable complexes with metal ions which are uni- or divalent and which are readily
deformed, this suggests that the metal ion involved falls into Pearson's class of 'soft' acids
(Pearson, 1966), which includes metals such as zinc and copper.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
Inhibitors o f influenza R N A polymerase
69
(.)
"I
Fig. 4. Formation of I : I complex by chelation of metal ion (M) with bathophenanthroline.
Bathophenanthroline, demonstrated above to be an effective inhibitor of influenza virusassociated RNA polymerase activity, chelates preferentially heavy-metal ions such as
copper, iron and zinc, forming a series of complexes such as the mono complexes illustrated
in Fig. 4 (Schiff, I969). Because of stereochemical considerations, substitution of positions
2 and 9 by methyl groups increases the stability constants of the bis-I,Io-phenanthrolinecopper(I) complex by IO aa, whereas for copper(II) and zinc(II) there are decreases by
factors of Io a7 and IO 4'a, respectively (Sillen & Martell, ~964, I97~). Although the corresponding stability constants are not available for bathophenanthroline and bathocuproine,
similar differences might be expected. That the difference in virus RNA polymerase enzyme
inhibition by bathophenanthroline and bathocuproine is marginal is more consistent with the
formation of a mixed enzyme-metal-ligand complex in situ than it is with abstraction of a
metal from the enzyme and the formation of a bis or a tris complex. In addition, studies
using atomic absorption spectrometry have detected zinc in preparations of purified
influenza A and B viruses and the zinc is not removed by dialysis against bathophenanthroline (J. S. Oxford & D. D. Perrin, unpublished observations).
Zinc and copper ions added to the influenza virus RNA polymerase enzyme system
showed inhibitory activity, and the inhibitory effect of adding bathocuproine-copper(I) and
bathocuproine-zinc complexes was greater than the effect of bathocuproine itself. It seems
reasonable to suggest that the ligand-copper or zinc complexes dissociate in solution so that
the copper or zinc is taken up by proteins, including the polymerase polypeptide(s), and
exert a separate inhibiting effect on the RNA polymerase enzyme. Both copper and zinc ions
are relatively potent inhibitors of a number of enzymes as well as the RNA polymerase
enzyme noted above (Perrin, ~97o). Also, at the same time, the ligand is made available to
chelate the hypothesized active metal of the polymerase enzyme polypeptide(s) exerting an
additional inhibitory effect on virus R N A polymerase activity. Levinson et al. 0973) have
reported the inhibition by heterocyclic thiosemicarbazones of the reverse transcriptase of
Rous sarcoma virus but suggest that the inhibitory effects are caused by the formation of a
ligand-metal complex. The multiplication of influenza A virus is inhibited in vitro by
certain metal chelates of substituted I,~o-phenanthrolines (White, Harris & Schulman,
I963). However, we consider that in the present studies if inhibition was due directly to
complexes formed between the chelating agent and either zinc or copper present as contaminants in the virus preparation or in the reagents of the RNA polymerase reaction mixture,
concentration quenching (Albert, ~968) would be expected; that is, with increasing ligand
concentration the dose-response curve would pass through a maximum, usually at less than
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
70
J.S.
OXFORD
AND
D.
D. PERRIN
Ioo ~ inhibition of enzyme activity, and then decrease. This behaviour was not observed.
At present, the main criticism of the hypothesis that the chelating agents are inhibiting
influenza virus RNA polymerase enzyme activity by chelating in situ to an essential metal of
the enzyme, such as zinc, is the failure to prevent or reverse the inhibitory effects of the
chelating agents by the addition of excess zinc to the enzyme reaction mixture. Inhibition of
the zinc metalloenzymes bovine liver glutamic hydrogenase, carboxypeptidase of bovine
pancreatic juice and lactic dehydrogenase of rabbit skeletal muscle by chelating agents can
be reversed by the addition of excess Zn(II) to the enzyme system (Vallee, I96o). Once a
metalloenzyme-inhibitor complex has been formed a critical excess of metal ion (Me) might
be expected to reverse the inhibition by competition with the metalloenzyme (EMe) for
bound inhibitor (In):
(EMe) In + Me ~- (EMe) + (Me In).
However, as discussed above, since zinc ions alone inhibit the influenza RNA polymerase
activity, the stoichiometry of the reactants is of importance, and it is possible that to date
the exact conditions for reversal have not been achieved. Studies with purified RNAdependent RNA polymerase enzyme and exogenous influenza RNA template will be
required to establish unequivocally the mode of action of these inhibitors. The isolation and
purification of influenza RNA polymerase will also allow direct determination of any heavy
metal.
Even compounds that inhibit completely the virus RNA polymerase enzyme in vitro may
have no effect on influenza virus multiplication in vivo and at present we have no information
on any virustatic effect of the chelating agents in tissue culture or in experimental animals.
However, if the RNA-dependent RNA polymerase enzyme of influenza virus is a metalloenzyme or heavy metal-activated enzyme, more effective and selective chelators of the metal
concerned could be synthesized. Such structure-activity studies then could be expected to
provide a more rational basis for the development of chemoprophylactic agents against
influenza virus. The principles involved in developing more effective compounds for chelation therapy are well understood, and have already been applied in designing a wide range
of chelating agents for use in treating metal ion toxicities (Perrin, I97o).
We would like to thank Ms Gertraud Stange for excellent technical assistance. Dr D. J.
Bauer, The WeUcome Research Laboratories, Beckenham, Kent, kindly supplied Nmethylisatin 3-thiosemicarbazone.
REFERENCES
ALBERT, A. (I968). In Selective Toxicity, 4 t h ed., p. 332. L o n d o n : M e t h u e n .
BAUER, D. J. (I965). Clinical experiments with the antiviral d r u g M a r b o r a n 0 - m e t h y l i s a t i n 3-thiosemicarbazone). Annals of the New York Academy of Sciences x3o, I I o--I 17.
BISHOP, D. H. L., ROY, V., BEAN, W. J. & SIMPSON, R. W. (I972). T r a n s c r i p t i o n of the influenza ribonucleic acid
g e n o m e by a virion polymerase. Ill. Completeness of the transcription process. Journal of Virology xo,
689-697.
CHOW, N. L. & SIMPSOr~, R. W. (~97 I). R N A d e p e n d e n t R N A polymerase activity associated with virions a n d
subviral particles of myxoviruses. Proceedings of the National Academy of Sciences of the United States
of America 68, 752-756.
rAZEr~AS DE ST GROTH, S. & WHITE, O. O. (I958). A n improved assay for the infectivity of influenza viruses.
Journal of Hygiene 56, I5I-x62.
GINGRAS, B. A., SUPRUNCHUK, T. & BAYLEY, C. H.
096z). The preparation
of some thiosemicarbazones and
their copper complexes. Part llI. Canadian Journal of Chemistry 4o, xo53-Io59.
GONICK, E. 0 9 5 0 - P h . D . Dissertation, Pennsylvania State College.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18
Inhibitors of influenza RNA polymerase
71
GONICK, E., FERNELIUS, W. C. & DOUGLAS, B. E. (1954). Equilibrium c o n s t a n t s for t h e f o r m a t i o n o f complexes
between metal ions and amines containing sulfur. Journal of the American Chemical Society 76 , 46714674.
HALL, J. R., MARCHANT, N. K. & PLOWMAN, R. A. (1962). C o o r d i n a t i o n c o m p o u n d s o f s u b s t i t u t e d l , i O - p h e n -
anthrolines and related dipyridyls. III. Complexes of copper and 2,9-dimethyl-I,IO-phenanthroline.
Australian Journal of Chemistry x5, 480.
HAWKINS, C. J. & PERRIN, D. D. (I963). Polynuelear complex formation. II. Cu(II) with cystine and related
ligands. Inorganic Chemistry 2, 843-849.
HO, P. P. K. & WALTERS, C. P. (I971). Inhibitors of influenza virus-induced R N A polymerase. Annals of the
New York Academy of Science i73 , 438-443.
HO, P. P. K., WALTERS, C. P., STREIGHTOFF, F., BAKER, L. A. & DELONG, D. C. (1967). In vivo activity of an
inhibitor o f influenza virus induced ribonucleic acid polymerase. Antimicrobial Agents and Chemotherapy 636-641.
LAVER, W. G. (1969). Purification of influenza virus. In Fundamental Techniques in Virology, p. 82. Edited by
K. Habel and N. P. Salzman. N e w York: Academic Press.
LEVINSON, W., FARAS, A., WOODSON, B., JACKSON, J. & BISHOP, J. M. (1973). I n h i b i t i o n o f R N A d e p e n d e n t D N A
polymerase of Rous sarcoma virus by thiosemicarbazones and several cations. Proceedings of the
National Academy of Sciences of the United States of America 7 o, 164-168.
MANN, F. G. & SAUNDERS, B. C. (1960). I n
Practical Organic Chemistry, p. 256. L o n d o n : L o n g m a n s .
OXFORD, J. S. (1973 a). A n inhibitor of the particle associated R N A dependent R N A polymerase of influenza
A and B viruses. Journal of General Virology 18, I1-19.
OXFORD, J. S. (I973b). Polypeptide composition o f influenza B viruses and enzymes associated with the
purified virus particles. Journal of Virology x2, 827-835.
PEARSON, R. G. (1966). Acids and bases. Science, New York xSI, 172-I77.
PERRIN, D. O. (I970)- In Masking and Demasking of Chemical Reactions, p. I83. New Y o r k : Wiley-Interscience.
SADLER, P. W. (1963). Chemotherapy of viral diseases. Pharmacological Reviews I5, 407-447.
SCHIFF, A. A. (I969). In Analytical Application of 1, I o-phenanthroline and related compounds, p. 59. Oxford:
Pergamon Press.
SCRUTTON, M. C., WU, C. W. & GOLDTHWAIT, D. A. (I97I). The presence and possible role of zinc in R N A
polymerase obtained from E. coll. Proceedings of the National Academy of Sciences of the United States
of America 68, 2497-25Ol.
SILLEN, L. G. & MARTELL, A. E. (I964). Stability constants. The Chemical Society Special Publication, no. 17.
SILLEN, L. G. & MARTELL, A. E. ( I 9 7 0 . Stability constants. The Chemical Society Special Publication, no 25.
Supplement no. i.
SKEHEL, S. J. (I97I). R N A dependent R N A polymerase activity of the influenza virus. Virology 45, 793-796.
SLATER, :I. P., MILDVAN, A. S. & LOEB, L. A. (1971). Zinc in D N A polymerases. Biochemical and Biophysical
Research Communication 44, 37-43.
VALLEE, B. L. (1960). Metal and enzyme interactions: correlation of composition, function and structure. In
The Enzymes, vol. Itt, p. 225. Edited by P. D. Buyer, H. Lardy and K. Myrbach. New York: Academic
Press.
WARREN, L. (1959). The thiobarbituric acid assay o f sialic acids. Journal of Biological Chemistry 2:34, I9711975.
WHITE, D. O., HARRIS, A.W. & SCHULMAN,A. (1963). Actions o f metal chelates o f substituted J, to-phenanthrolines on viruses and cells, iI. Inhibition of viral multiplication. Australian Journal of Experimental
Biology 4I, 527-538.
WILLIAMS, D. R. (197I). The metals o f life. Ill The Solution Chemistry of Metal Ions in Biologieal Systems.
L o n d o n : Van Nostrand Reinhold Company.
(Received z 8 August ~973)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 22:09:18