- Email: [email protected]
The effects of S-Adenosyl methionine (AdoMet) and its analogues on the control of transcription and translation in vitro of the mRNA products of two cytoplasmic polyhedrosis viruses
VIROLOGY
131, 18-a
(1983)
The Effects of S-Adenosyl Methionine (AdoMet) and Its Analogues on the Control of Transcription and Translation in Vitro of the mRNA Products of Two Cytoplasmic Polyhedrosis Viruses P. P. C. MERTENS’ NERC Institute
of Virology,
Mansjield
Received Ap.1
AND Road,
C. C. PAYNE’ Oxfd OX1 SSR, Unit& Kin&am
7, 198s; accepted July 21, 1983
S-Adenosyl methionine (AdoMet) and several structurally related compounds were added to in vitro systems for the synthesis of single-stranded RNA by cytoplasmic polyhedrosis virus (CPV) types 1 and 2. The effects of these compounds were examined on the level of transcription and methylation of the RNA products. Of the compounds tested, five increased the polymerase activity in both viruses, the most effective being the D- and L-stereoisomers of S-adenosyl homocysteine (AdoHcy), and the least effective, adenosine. L-AdoHcy, unlike rrAdoHey, was also a competitive inhibitor of RNA methylation in the presence of PHjAdoMet. The different response of both viruses to D- and L-AdoHcy suggests that CPV virions contain at least two functionally distinct sites to which AdoMet, or its analogues, bind. One of these is the transcription control site, while the other is the active site(s) for RNA methylation. CPV RNA synthesised in the presence of the methyl donor AdoMet was more efficiently translated in vitro in a wheat-germ translation system than RNA synthesised in the presence of methylation inhibitors. Type 2 CPV-RNA transcripts had a greater degree of methylation than type 1 CPV transcripts and were more effective in stimulating protein synthesis in the translation system. It seems likely that the allosteric control of CPV polymerase by AdoMet and its analogues, and the methylation of the transcripts, ensures the effective transcription and translation of the CPV genome and the stability of the viral messenger RNA.
also stimulate the polymerase activity of this virus although several of them, ineluding S-adenosyl homocysteine (AdoHcy) cannot serve as methyl group donors (Mertens and Payne, 1978; Furuichi, 1978a; Wertheimer et aL, 1980). These results suggest that transcription in type 1 CPV is not directly linked to RNA methylation. Virions of type 2 CPV will also synthesise and methylate ssRNA transcripts of all 10 viral RNA genome segments in vitro. However, with this virus tvpe there is onlv a partial dependence on the presence of AdoMet, or structurally related compounds, for efficient ssRNA synthesis (Mertens and Payne, 1978). Recent studies have shown that the stimulatory effect of AdoMet is governed by some interaction between it and the virion enzymes which leads to an allosteric modification and results in an increase in
INTRODUCTION
S-Adenosyl methionine (AdoMet) acts as a methyl group donor during the in vitro synthesis of “capped” messenger RNA (mRNA) by the virion-associated polymerase of cytoplasmic polyhedrosis viruses (CPVs) (Furuichi, 1974, 1978a, 1981; Mertens and Payne, 1978). In addition, AdoMet massively stimulates RNA transcription by the polymerase of type 1 CPV. Other compounds related in structure to AdoMet can
’ Present address: Department of Biochemistry Animal Virus Research Institute, Pirbright, Woking, Surrey GU24 ONF’, United Kingdom. Author to whom requests for reprints should be addressed. * Present address: Department of Entomology and Insect Pathology, Glasshouse Crops Research Institute, Worthing Road, Littlehampton, West Sussex BN16 3P1J, United Kingdom. 0042-6822/83 fLWCI Copyriyht All rights
G 19E!3 hy Academic Prws, Inc. of reproduction in any form reserved.
18
CPV mRNA
SYNTHESIS
AND
polymerase activity (Furuichi, 1981). It has been suggested that AdoMet or its analogues bind to the active site of the methyl transferase(s) in the virions. In so doing, they might not only serve as methyl donors or competitive inhibitors of methylation but also stimulate RNA synthesis (Wertheimer d al, 1980). In this paper we present results which are inconsistent with this interpretation. We suggest that the effects of AdoMet or its analogues are due to their binding to at least two functionally distinct sites on the virus particle; the transcription control site and the methyl transfer site(s). In the present study, several compounds structurally related to AdoMet were examined for their effects on the polymerase activities of CPV types 1 and 2. The main compounds used were AdoMet, the D- and L-stereoisomers of AdoHcy (D-AdoHcy and r..-AdoHcy, respectively), S-adenosylL-ethionine (AdoEth) and adenosine. An analysis was also made of the ability of viral mRNA synthesized in vitro (in the presence of each of these compounds) to stimulate protein synthesis in a wheatgerm translation system. The results supported the conclusion that effective protein synthesis in vitro is dependent on the presence of complete methylated “cap” structures at the 5’ terminus of CPV mRNA molecules. MATERIALS
AND
METHODS
Viruses. The preparation of polyhedra and virus particles of CPV types 1 and 2 for these experiments, was identical to the methods described in a previous publication (Mertens and Payne, 1978). Electrophoretic analysis of RNA. Singlestranded (ss)RNA synthesised in the polymerase assay system was analyzed by electrophoresis using a 1.8% polyacrylamide, 0.6% agarose, 6 M urea gel system (Schuerch et &, 1975). After electrophoresis, the gels were stained with 0.01% ethidium bromide in 0.02 M acetate buffer (pH 7.8) and the ssRNA bands were visualised and photographed under ultraviolet illumination. Some gels containing “dual-labelled” ssRNA were sliced into l-mm slices, which were then treated with 200 ~1 of 60% (w/v) hydrogen peroxide overnight at 60”.
TRANSLATION
IN VZTRO
19
The dissolved slices were resuspended in 10 ml of a scintillation cocktail containing 66.7% toluene, 33.3% Triton X-100, 0.4% (w/v) 2,5-diphenyl-oxazole (PPO), and 0.05% (w/v) 1,4-di-2-(5-phenyl-oxazole) benzene (POPOP). The amount of “H and “P in each slice was determined in a Packard-Tricarb liquid scintillation spectrometer. Polymerase assays, Polymerase assay conditions were as previously described (Mertens and Payne, 1978). Structural analogues of AdoMet were included at concentrations indicated in the text. All analogues were obtained from the Sigma Chemical Co., Ltd. For the production of unlabelled ssRNA the standard polymerase assay was scaled up X30; radioactive UTP was omitted from the reaction system and the “cold” UTP concentration was increased to 2.0 mM. After incubation, virus particles and bentonite were removed by centrifugation (70,000 g for 1 hr at 4”), the supernatant was extracted with an equal volume of water-saturated phenol to remove any traces of ribonuclease, and then washed three times with diethyl ether. The ssRNA was precipitated by the addition of 0.25 vol of 10.0 M lithium chloride (sterile, final concentration, 2.0 M), followed by storage for 16 hr at 0” (Furuichi, 1974). The precipitate was pelleted by centrifugation (30,000 g, 10 min at 4“), washed twice with ethanol and once with diethyl ether, dried, resuspended in a small volume of sterile deionized water, and then stored frozen at -20”. The concentrations of these preparations of ssRNA were determined from their absorbance at 260 nm. Cell free translation assays. A cell free translation system was produced from wheat germ using the experimental procedure described by Roberts and Paterson (1973). The wheat-germ extract was stored as lOO-j.J aliquots in sterile glass vials under liquid nitrogen and transfer (t)RNA was not added to this assay system. The translation assays had a final volume of 25 ~1, containing 10 ~1 of wheat-germ extract, 20 mMHEPES buffer (pH 7.6), 2 mM dithiothreitol, 1 mM ATP, 20 pM GTP, 8 mM creatin phosphate, 40 pg/ml of creatin phosphokinase, 20-30 pM of the appropriate amino acids, 80 mM potassium chloride,
20
MERTENS
AND
2.5 mM magnesium acetate, 5 PCi of [3H]leucine, and CPV ssRNA as indicated in the text. Reaction mixtures were incubated for 1 hr at 25”. The reaction was stopped by the addition of 200 ~1 of 0.1 M KOH, followed by incubation at 37“ (to hydrolyse amino-acylated tRNA molecules). The level of protein synthesis was determined by measuring the amount of label incorporated into material precipitated by the addition of 10 ml of 10% TCA. After storage for 1 hr at 4”, the precipitate was collected on glass fiber filters, washed with 40 ml of 10% TCA, 10 ml of ethanol, and then dried. The filters were placed in 5 ml of a toluene-based scintillation cocktail containing 0.4% (w/v) PPO, 0.05% (w/v) POPOP, and then counted in a PackardTricarb liquid scintillation spectrometer. Translation assays were performed in duplicate or triplicate for each set of reaction conditions (as indicated in the text) and the average result for amino acid incorporation was calculated. RESULTS
Polymer-me Activity AdoMet, L-AdoHcy, D-AdoHcy, and AdoEth all stimulate the polymerase activity of CPV types 1 (Fig. la) and 2 (Table
PAYNE
1, Fig. lb). Of the nucleosides, nucleoside monophosphates and nucleoside diphosphates tested only adenosine increased ssRNA synthesis (Table 1). There was no observable stimulation of transcription caused by either homocysteine or methionine in this system (Table 1, Fig. 1). The addition of increasing concentrations of the stimulating compounds to reaction mixtures containing virions of either CPV type increased ssRNA synthesis up to a maximum level (Fig. 1). Beyond this level little or no increase of transcription was caused by the further addition of any of the other stimulation compounds tested (Table Z), suggesting that these compounds have a similar mode of action. When analysed at low concentrations, the most effective stimulator for either CPV was L-AdoHcy and the least effective was adenosine (Fig. 1). RNA Analysis Single-stranded RNAs synthesised in the presence of 0.5 mM AdoMet, L-AdoHcy, DAdoHcy, AdoEth, or 1.5 mM adenosine, were analysed by gel electrophoresis. The relative amounts of RNA synthesised (measured by UMP incorporation) in the presence of these compounds are shown in
c mMolar
concentration
of adenosine
analogue
FIG. I. A comparison of the effect on polymerase activity produced by the addition concentrations of AdoMet (0), I.-AdoHcy (O), AdoEth (Cl), D-AdoHey (A), Adenosine mocysteine (A) to assays containing virions of (A) type 1 and (B) type 2 CPV.
of varying (W, or ho-
CPV mRNA
SYNTHESIS
AND TABLE
TRANSLATION
21
IN VITRO
1
EFFECT OF A RANGE OF CHEMICALS ON CPV POLYMERASE Acme Nanomoies Addition to standard assay0
Molarity (mW
No addition AdoMet AdoMetb AdoMet L-AdoHcy * D- AdoHcy b AdoEthb
of UMP incorporated
Type 1 CPV (40 &ml)
Type 2 CPV (40 mz/ml)
0.05 0.5 1.5 0.5 0.5 0.5
0.33 7.25 7.36 8.81 8.10 8.25 8.30
1.80 4.21 5.27 4.01 5.40 5.51 5.60
0.13 0.10 0.13 0.12 0.11 0.10 0.11
AMP CMP GMP UMP
0.5 0.5 0.5 0.5
0.34 0.33 0.36 0.34
1.80 1.61 1.50 1.73
0.14 0.10 0.11 0.13
ADP CDP GDP UDP
0.5 0.5 0.5 0.5
0.35 0.33 0.32 0.34
1.64 1.60 1.71 1.75
0.11 0.11 0.13 0.14
Adenosine Adenosineb Cytidine Guanosine IJridine
0.5 1.5 0.5 0.5 0.5
0.93 3.87 0.34 0.34 0.35
3.26 4.20 1.62 1.81 1.73
0.13 0.15 0.11 0.10 0.14
Methionine Homocysteine
0.5 0.5
0.33 0.35
1.54 1.72
0.10 0.10
a Assays were incubated for 5 hr at 31” and were performed in triplicate. bssRNA synthesised in large-scale assays under these concentrations of stimulators elcctrophoresis (Fig. 2).
Table 1. The electrophoretic profiles of the products from the two viruses remained practically constant regardless both of the nature of the stimulator used (Fig. 2), or of different concentrations of AdoMet (0.05-1.5 mkf; results not shown). These experiments clearly demonstrate that although the specific nature and concentration of the stimulator used affect the overall level of transcription by these CPVs they have little effect on the relative levels of transcription of the different genome segments. Each RNA segment of both viruses is transcribed in approximately equal amounts by weight (Smith and Furuichi, 1980; Payne and Mertens, 1983).
Control (no virus)
were analyzed
by
RNA Methyl&cm A modified (dual label) polymerase assay system, containing [w~P]UTP to measure RNA synthesis and [methyl-3H]AdoMet to measure methylation, was used to analyse the effects caused by L-AdoHcy, D-AdoHcy, AdoEth, or adenosine on methylation of the polymerase products of CPV types 1 and 2. Using this system, it was found, as previously reported (Mertens and Payne, 1978; Furuichi, 1978a), that AdoHcy was an effective inhibitor of the methyl transferase activity of both CPV types (Figs. 3a, b, e, f). AdoEth also inhibited transmethylation (Figs. 3c, g) and appeared to be
22
MERTENS AND PAYNE TABLE2 COMPARISON OF THE STIMULATION OF CPV POLYMERASE ACWITY BY AdoMet, D-AdoHcy, AdoEth, AND ADENOSINE, ADDED SEPARATELY OR TIXETHER Nanomoles
Addition
to standard
No addition AdoMet, 0.5 mM L-AdoHcy, 0.5 m&f AdoEth, 0.5 mM D-AdoHcy, 0.5 mM Adenosine, 1.5 mM AdoMet, 0.5 mM t AdoMet, 0.5 m&f + AdoMet, 0.5 m&f t AdoMet, 0.5 m&f t L-AdoHcy, 0.5 mM L-AdoHcy, 0.5 mM L-AdoHcy, 0.5 mM D-AdoHcy, 0.5 mM D-AdoHcy, 0.5 mM AdoEth, 0.5 m&f + ‘Assays
were incubated
assay system’
LXst~tim
An analysis was made of the methyl distribution between the different size classes of the polymerase products separated by electrophoresis (Fig. 4). Although the definition achieved by the slicing of these gels was not sufficiently good to permit an estimate of the levels of methylation of each of the product species;
group
Type 2 CPV (40 &ml)
0.5 a.4 8.7 9.6 9.2 3.8 9.4 9.5 9.0 8.9 9.2 9.1 9.2 9.8 9.3 9.4
2.4 4.4 4.4 4.7 4.8 4.2 4.3 3.7 4.7 3.8 4.0 4.4 4.1 4.8 4.7 4.8
for 5 hr at 32.0” and were performed
of Methyl Groups
of UMP incorporated
Type 1 CPV (40 &ml)
L-AdoHcy, 0.5 mM AdoEth, 0.5 mA4 D-AdoHcy, 0.5 mM adenosine 1.5 mM t AdoEth, 0.5 mM t D-AdoHcy, 0.5 mM + adenosine, 1.5 mM + AdoEth, 0.5 mM + adenosine, 1.5 mM adenosine, 1.5 mM
slightly more effective in this respect (at lower concentrations) than I.-AdoHcy (Figs. 3b, f). The addition of AdoEth or LAdoHcy did not, however, cause any observable inhibition of methyl group incorporation until they were added to these assays at concentrations of approximately two to four times that of the [methyl‘H]AdoMet already present. In contrast, D-AdoHcy (a more effective stimulator of ssRNA synthesis than AdoMet) and adenosine (the least effective stimulator) did not greatly reduce methylation of the polymerase products of either type 1 (Figs. 3a, d) or type 2 (Figs. 3e, h) CPV at the concentrations tested.
L-AdoHcy,
in triplicate.
it is clear that, as with reovirus ssRNA (Levin and Samuel, 1977), the incorporated methyl groups were not restricted to the transcripts of any particular genome segment(s) but appeared to be distributed between the different size classes separated by this technique.
Translaticm of ssRNA The nature of the cap structure present at the 5’ termini of the ssRNA synthesised by these CPV in vitro, is affected by the nature of the stimulators present in the polymerase assay system (Furuichi, 1978a, b; Mertens and Payne, 1978, Wertheimer et al, 1980). The cap structures and the degree of methylation may well, in turn, affect the stability and messenger activity of these ssRNA molecules (Furuichi and Miura, 1975; Muthukrishnan et al, 1975; Shih et ah, 1976; Levin and Samuel, 1977; Shimotohno et aL, 1977, Furuichi, 1978a; Banerjee, 1980). Single-stranded RNA synthesised by CPV types 1 and 2 in the presence of each of the five stimulators tested (Fig. 2) was
CPV mRNA
SYNTHESIS
AND
TRANSLATION
IN
VITRO
Type 1
FIG. 2. Electrophoretic analysis of ssRNA synthesised in large-scale polymerase assays by CPV types 1 and 2 in the presence of 0.5 mML-AdoHcy (a), 0.5 mM AdoMet (b), 0.5 ~MD-AdoHcy (c), 0.5 mM AdoEth (d), or 1.5 mM adenosine (e). The same amounts of RNA were analysed on each gel despite differences in the levels of polymeraae activity under different reaction conditions (Table 1). The gels were photographed after staining with ethidium bromide.
adcled at varying concentrations to wheatassays. The messenger gel mm translation act ivity of each preparation of RNA was asslessed from its ability to increase amino aci d incorporation in this system (Fig. 5).
The preparations of “unmethylated” ssRNA synthesized by both viruses in the presence of L-AdoHcy, D-AdoHcy, or adenosine all produced very similar “stimulation curves” when tested in the wheat-germ system.
24
MERTENS Type
AND
PAYNE
1 CPV
Type 2 CPV
e 0 .
/@-H-O D-Ado
Hey
L b
f
a-0
\
L-Ado
HCY
l -.-+-*-------. Ado Elh
Adenosine
81
O-04
r-•-*-
g
d
-a
u
OLI
I
J
I
0
0.5
1.0
0
.
l -H-’
e-
Adenosine I 0.5
I 1.0
FIG. 3. A comparison of the effects of varying concentrations of (a and e) D-AdoHcy, (b and f) L-AdoHcy; (c and g). AdoEth; (d and h) adenosine on methyl group (0) and UMP (0) incorporation by virions of type 1 CPV (a-d) and type2 CPV (e-h) under standard “dual-label” reaction conditions, using [cI-~P]UTP and [%iJAdoMet.
4,300
6,500
A
750
900 r r*
.
VW7
600
0 0
80
0
80
Gel - length (mm) FIG. 4. Single-stranded RNAs of (A) type 1 and (B) type 2 CPV, dual-labclled using [(u-~P]UTP and [‘H]AdoMet, were analysed by electrophoresis. After staining, the gels were cut into l.O-mm slices and the amounts of =P and ‘H in each slice were determined. The position of the bands which were visible after staining are marked by arrows.
CPV mRNA
SYNTHESIS
AND
TRANSLATION
Zhr VZTRO
A
pg of RNA added per assay FIG. 5. The messenger activity of CPV ssRNA, synthesised in the presence of 0.5 mM AdoMet (O), 0.5 mM I.-AdoHcy (O), 0.5 mM D-AdoHcy (A), 0.5 mM AdoEth (Cl), and 1.5 mM adcnosine (m) in wheat-germ translation assays. The preparations of RNA used were the same as those analysed in Fig. 2. (A) Type 1 CPV RNA; (B) type 2 CPV RNA.
“Methylated” ssRNA, synthesised in the presence of AdoMet increased amino acid incorporation considerably. This difference was particularly noticeable with the preparations of ssRNA from type 2 CPV (Fig. 5b), possibly because the average level of methylation in the products from the type 2 CPV polymerase is higher than in those from type 1 CPV (see Discussion). RNA synthesised by either CPV type in the presence of AdoEth, caused a lower level of stimulation than the ssRNA synthesised in the presence of any of the other four transcription stimulators tested. Some capping and methylation of ssRNA may occur in the wheat-germ assay system (Levin and Samuel, 1977). It was therefore possible that the differences in messenger activity of the different preparations of CPV ssRNA were at least partially masked under standard translation assay conditions. In view of this possibility, an analysis was made of the effects produced by the addition of AdoMet, L-AdoHcy (a stimulator and inhibitor of transmethylation, respectively), and the other CPV transcription stimulators (D-AdoHcy, AdoEth, and adenosine) to the wheat-germ assay system. The results obtained using “un-
methylated” RNA synthesised in the presence of D-AdoHcy and adenosine were similar to those obtained with L-AdoHcy and are not shown in Fig. 6. The addition of AdoMet to the translation assays increased amino acid incorporation in response to “unmethylated” ssRNA of either CPV type (synthesised in the presence of L-AdoHcy; Figs. 6b, e). Some increase in amino acid incorporation was also observed when AdoMet was added to assays containing “methylated” ssRNA (synthesised in the presence of AdoMet; Figs. 6a, d). However, the overall increase, particularly with ssRNA from type 2 CPV (which has a higher level of methylation) was considerably less than with unmethylated material (compare Fig. 6d with Fig. 6e). When AdoMet was added to translation assays containing ssRNA synthesised in the presence of AdoEth, there was only a small increase in amino acid incorporation comparable to that produced in the presence of methylated ssRNA (Figs. 6c, f). This may well be due to the incorporation of ethyl groups into the RNA synthesised in the presence of AdoEth (Furuichi, 1978a, b), preventing their subsequent methyl-
26
MERTENS Type
AND
PAYNE
1 CPV
Type 2 CPV a
0
0.5
1.0
0
0.5
m~olarconcentrationof adenosineanalogue FIG. 6. The effects of the addition of AdoMet (0), r.-AdoHcy (O), D-AdoHcy (A), AdoEth (O), or adenosine (W) to wheat-germ translation assays containing (a and d) “mcthylated” CPV ssRNA (synthesised in the presence of AdoMet), (b and e) “unmethylated” ssRNA (synthesised in the presence of L-AdoHcy); and (c and f) “ethylated” ssRNA (synthesised in the presence of AdoEth).
ation. None of the other compounds tested (L-AdoHcy, AdoEth, D-AdoHcy, or adenosine) increased amino acid incorporation when added to the wheat-germ system. DAdoHcy appeared to have little effect on the level of protein synthesis, regardless of the ssRNA preparation used. Adenosine inhibited protein synthesis and at the highest concentration tested (1.0 mM), reduced amino acid incorporation to near background (endogenous activity). The remaining compounds, L-AdoHcy and AdoEth which both inhibited transmethylation reactions in the polymerase assay system inhibited protein synthesis to an almost identical extent when added to the wheat-germ assay system. This effect, like the stimulation caused by AdoMet was most pronounced in assays containing “unmethylated” CPV ssRNA (Figs. 6b, e).
DISCUSSION
The original discovery that the methyl donor, AdoMet, increased ssRNA synthesis by type 1 CPV polymerase first suggested that transcription and methylation of the RNA were linked (Furuichi, 1974). That this was not the case was clearly demonstrated by further studies using analogues of AdoMet which are not methyl donors but which, nonetheless, increase polymerase activity (Mertens and Payne, 1978; Furuichi, 19’78a;Wertheimer et aL, 1980). In the present study, we have compared the effects of AdoMet and structurally related compounds on the transcription and translation of CPV ssRNAs. Similar relative levels of the ssRNA transcripts of all the viral genome segments were produced regardless of the different molecular
CPV mRNA SYNTHESIS AND TRANSLATION IN VITRO structures of the compounds which stimulated polymerase activity. In addition, all these compounds appeared to increase transcription by a basically similar mechanism as the “maximum” level of polymerase activity could not be increased when AdoMet and its analogues were added in combination, rather than singly. Perhaps of greatest interest were the differential effects of these compounds on RNA methylation; in particular, the different responses of the viruses to the stereoisomers, D- and L-AdoHcy. Both compounds increased polymerase activity but, while L-AdoHcy inhibited methylation, DAdoHcy did not greatly reduce methyl group incorporation into the RNA transcripts. Previous studies have suggested that the most likely explanation for the effect of compounds such as AdoMet or AdoHcy, is that they bind to a single site on the virus and both stimulate RNA synthesis and serve as methyl donors or competitive inhibitors of methylation (Wertheimer et a,l., 1980). D-AdoHcy does not fit into this pattern. Nor do other compounds which inhibit CPV-mRNA synthesis in vitro without reducing methylation. These include compounds such as S-guanosyl-L-methionine(GuaMet),S-uridyl-L-methionine(UriMet), and AdoHcy-dialdehyde (Wertheimer et cd., 1980). However, these anomalies can be interpreted by an alternative model (also considered by Wertheimer et al, 1980) in which molecules of AdoMet or its analogues bind separately to functionally distinct sites, i.e., one to a methyl transfer site(s) (Payne and Mertens, 1983), the other to a transcription control site. While AdoMet and L-AdoHcy may bind efficiently to both sites, DAdoHcy, GuaMet, UriMet, and AdoHcydialdehyde may only bind to the site involved in transcription control. In the latter group of compounds, only the binding of D-AdoHcy leads to increased transcription. In contrast to type 1 CPV, the transcription control site in type 2 CPV appears to be under less rigid control by AdoMet or its analogues, as the virion polymerase is relatively active in their absence. The different responses of the polymerase of both CPV types 1 and 2 to AdoMet
27
and its analogues are useful in evaluating the structural features of the molecules which are important in the promotion of efficient transcription. Although adenosine increased polymerase activity, the most effective compounds tested by us, bore an amino acid side chain. Differences in the nature and conformation of the side chain (including the sulphonium group) gave minor differences in polymerase activity, although the amino acids homocysteine or methionine by themselves did not stimulate transcription. These results are in general agreement with those of Wertheimer et al. (1980), and suggest that no single structural feature has yet been defined as the key component in CPV polymerase stimulation. However, it is interesting that in a recent study, Wu et nl. (1981) have shown that methyl-methionine alone is also very effective in increasing polymerase activity in type 1 CPV virions. Studies on the methylation of CPV-RNA showed, as with reovirus (Levin and Samuel, 19’7’7) that methyl groups were not restricted to any specific genome segments. There does not, therefore, appear to be any control of the relative levels of expression of the different genome segments of CPV types 1 and 2, mediated either by the differential stimulation of transcription of a particular segment, or by an uneven distribution of methyl groups between transcripts. Assuming that the majority of the radioactive substrates incorporated into TCA precipitable material in the dual-label experiments represented macromolecular components of newly synthesised RNA and does not include the small oligonucleotides produced in the assay (Smith and Furuichi, 1982), it is possible to calculate the average number of methyl groups attached to each ssRNA chain. Taking into account the relative frequency of transcription of the different size classes of RNA (Payne and Mertens, 1983), the average MW of the ssRNA chains was 0.42 X 10’ and 0.52 X 10” for the products of CPV types 1 and 2, respectively. Taking the average uridine content as 28.5 and 31.5% for the RNA of CPV types 1 and 2, respectively (Miura et al, 1968; Payne and Churchill, 1977), it was calculated that type 2 CPV incorporated
28
MERTENS
an average of 1.0 methyl group per RNA molecule and type 1 CPV incorporated 0.52 groups per RNA molecule, under standard dual-label reaction conditions. These values are considerably lower than those published by us from similar experiments as it was incorrectly assumed that the segments of the CPV genome were all transcribed with approximately equal frequency (Mertens and Payne, 1978) rather than in equivalent amounts by weight (Smith and Furuichi, 1980; Payne and Mertens, 1983). From reports by other workers on the 5’ terminal structure of type 1 CPV RNA, it is probable that methyl groups are incorporated into a blocked or cap structure (m7GpppAmpGp . . .) at the 5’ termini of the polymerase assay products (Furuichi and Miura, 1975; Furuichi, 1978a; Shimotohno and Miura, 1974, 1976). The results of the analysis of the messenger activity of CPV ssRNA in the wheat-germ assay system are consistent with a requirement for methylation of the 5’ terminal cap structure for efficient initiation of translation (Levin and Samuel, 1977; Shimotohno et aL, 1977; Banerjee, 1980). The preparation of ssRNA which caused the highest specific stimulation of protein synthesis was that synthesised by type 2 CPV in the presence of AdoMet (“methylated” ssRNA). The ssRNA from type 1 CPV synthesised under similar reaction conditions caused a smaller increase in amino acid incorporation, possibly as a result of the lower level of methylation observed in this material. The “unmethylated” ssRNA preparations from both CPVs (synthesised in the presence of D-AdoHcy, L-AdoHcy, or adenosine) produced very similar stimulation curves, suggesting that in the “unmethylated” form (Furuichi, 1981), there is little intrinsic difference in the translational efficiency of ssRNA from these two viruses. The ssRNA synthesised by either CPV type in the presence of AdoEth produced the lowest levels of stimulation of those preparations tested. This may well result from the incorporation of ethyl groups at the 5’ terminal cap structure (Furuichi, 1978b) and suggests that ethylated cap structures block efficient translation in this system.
AND
PAYNE
In conclusion, the polymeraseand methylation-control mechanisms within the virions of CPVs do not appear to affect the relative levels of expression of the different CPV genome segments. It seems probable that these control mechanisms will help to ensure a high level of capping and methylation of all of the ssRNA products. However, if the end products of the transmethylase and polymerase reactions (AdoHcy and pyrophosphate, respectively) can accumulate during CPV replication, then it is possible that some progressive reduction could occur in the levels of capping and methylation at later stages of infection. Such changes could have considerable significance in affecting the processing and stability of the viral messenger RNA (Furuichi and Miura, 1975; Banerjee, 1980; Payne and Mertens, 1983). ACKNOWLEDGMENTS The authors would like to thank Dr N. F. Moore for his assistance in setting up the wheat-germ system and Dr K. A. Harrap for useful discussion of experimental procedures. REFERENCES BANEKJEE, A. K. (1980). 5’-terminal cap structure in eukaryotic messenger ribonucleic acids. Mirrohiol. Rev. 44, 175-205. FUKUI~HI, Y. (1974). Methylation-coupled transcription by virus associated transcriptase of cytoplasmic polyhedrosis virus containing double-stranded RNA. Nuu?eic Aa& Res. 1,809-822. FUKIJI~HI, Y. (1978a). Pre-transcriptional capping in the biosynt.hesis of cytoplasmic polyhedrosis virus mRNA. Proc Not. Acud Sci USA 75,1086-1990. FURUICIII, Y. (19’78b). Stimulation of cytoplasmic polyhedrosis virus mRNA synthesis by S-adenosylmethionine and its derivatives. “Abstracts of the Fourth International Congress for Virology, The Hague,” p. 332. Ccntre for Agricultural Publishing and Documentation, Wageningen. FURUICHI, Y. (1981). Allosteric stimulatory effect of S-adenosyl methionine on the RNA polymeraae in cytoplasmic polyhedrosis virus. A model for the positive control of eukaryotic transcription. J. BioL Chem 256, 483-493. FURUICHI, Y., and M~JRA, K. (1975). A blocked structure at the 5’ terminus of mRNA from cytoplasmic polyhedrosis virus. Nuture (Lmdm) 26.3, 374-375. LEV~N, K., and SAMUEL, C. E. (1977). Biosynthesis of reovirus specified polypeptides. Effect of methylation on efficiency of rcovirus genome expression in vitro. Virokgy 77, 245-259.
CPV mRNA
SYNTHESIS
AND
MERTENS, P. P. C., and PAYNE, C. C. (1978). S-adenosylL-homocysteine as a stimulator of viral RNA synthesis by two distinct cytoplasmic polyhedrosis viruses. .I. ViroL 26, 332-835. MIURA, K., FUJII, I., SAKAIU, T., FUKE, M., and KAWASE, S. (1968). Double-stranded ribonucleic acid from cytoplasmic polyhedrosis virus of the silkworm. J. ViroL 2, 1211-1222. MUTHIJKRISHNAX, S., BOTH, G. W., FURUICHI, Y., and SHATRIN, A. J. (1975). 5’ terminal ‘I-methyl guanosine in eukaryotic mRNA, required for translation. Nature &m&m) 255,3337. PAYNE, C. C., and CHURCHILL, M. P. (1977). The specificity of antibodies to dsRNA in antisera to three distinct cytoplasmic polyhedrosis viruses. Virology, 79, 251-258. PAYXE, C. C., and MERTENS, P. P. C. (1983). Cytoplasmic polyhedrosis viridae. In “The Reoviridae” (W. K. Joklik, ed.), pp. 425-504, Plenum, New York. ROBERTS, B. E., and PATERSON, B. M. (1973). Efficient translation of tobacco mosaic virus RNA and rabbit globin ts RNA in a ceil-free system from commercial wheat germ. Proc. Nat. AU& sci, USA 70, 2330‘2334. SCHUERCH, A. R., MIT~HGLL, W. R., and JOKLIK, W. K. (1975). Isolation of intact individual species of single and double-stranded RNA after fractionation by polyaerylamide gel electrophoresis. Anal. Biochem 65, 331-345. SIIIH, D. S., DASGUPTA, A. R., and KAESBERG, P. (1976). ‘I-methyl-guanosine and the efficiency of RNA translation. J. ViroL 19. 637-642
TRANSLATION
IN
VITRO
29
SHIMOTOHXO, IL, and Mums, K.-I. (1974). <5’-terminal structure of messenger RNA transcribed by the RNA polymerase of silkworm cytoplasmic polyhedrosis virus containing double-stranded RNA. J. MoL &id 86.21-30. SHIMOTOHNO, K., and MIVRA, K.-I. (1976). The process of formation of the S’terminai modified structure in messenger RNA of cytoplasmic polyhedrosis virus. FEBS Z&t 64.204-208. SHIYOTOHNO, K., KODAMA, Y., ~~~~~~~~~~~~ J., and MICRA, K.-I. (1977). Importance of g-terminal blocking structure to stabilize mRNA in cukaryotic protein synthesis. E’roc. Nat Acd Sci USA 74, 2734-2738. SMITH, R. E., and FUR~JI~I-II, Y. (1980). Gene mapping of cytoplasmic polyhcdrosis virus of silkworm by the full-length mRNA prepared under optimized conditions of transcription in vitro. Virology 103, 279-290. WERTHEIMER, A. M., CHEN, S. Y., BORCHARI)T, R. T., and FIJRUICHI, Y. (1980). S-adenosyl methionine and its analogs: Structural features correlated with synthesis and methylation of mRNAs of cytoplasmic polyhedrosis virus. J. BioL C&-m. 255,59245930. WV, A., DAI, R., SHEN, X., and SUN, Y. (1981). [‘IImethyl]-methionine as possible methyl donor for formation of ti’-terminus of in vitro synthesized mRNA of cytoplasmic polyhedrosis virus of silkworm, Bwntn~;c mot-i. Ski Sin 24,1737-1742.
131, 18-a
(1983)
The Effects of S-Adenosyl Methionine (AdoMet) and Its Analogues on the Control of Transcription and Translation in Vitro of the mRNA Products of Two Cytoplasmic Polyhedrosis Viruses P. P. C. MERTENS’ NERC Institute
of Virology,
Mansjield
Received Ap.1
AND Road,
C. C. PAYNE’ Oxfd OX1 SSR, Unit& Kin&am
7, 198s; accepted July 21, 1983
S-Adenosyl methionine (AdoMet) and several structurally related compounds were added to in vitro systems for the synthesis of single-stranded RNA by cytoplasmic polyhedrosis virus (CPV) types 1 and 2. The effects of these compounds were examined on the level of transcription and methylation of the RNA products. Of the compounds tested, five increased the polymerase activity in both viruses, the most effective being the D- and L-stereoisomers of S-adenosyl homocysteine (AdoHcy), and the least effective, adenosine. L-AdoHcy, unlike rrAdoHey, was also a competitive inhibitor of RNA methylation in the presence of PHjAdoMet. The different response of both viruses to D- and L-AdoHcy suggests that CPV virions contain at least two functionally distinct sites to which AdoMet, or its analogues, bind. One of these is the transcription control site, while the other is the active site(s) for RNA methylation. CPV RNA synthesised in the presence of the methyl donor AdoMet was more efficiently translated in vitro in a wheat-germ translation system than RNA synthesised in the presence of methylation inhibitors. Type 2 CPV-RNA transcripts had a greater degree of methylation than type 1 CPV transcripts and were more effective in stimulating protein synthesis in the translation system. It seems likely that the allosteric control of CPV polymerase by AdoMet and its analogues, and the methylation of the transcripts, ensures the effective transcription and translation of the CPV genome and the stability of the viral messenger RNA.
also stimulate the polymerase activity of this virus although several of them, ineluding S-adenosyl homocysteine (AdoHcy) cannot serve as methyl group donors (Mertens and Payne, 1978; Furuichi, 1978a; Wertheimer et aL, 1980). These results suggest that transcription in type 1 CPV is not directly linked to RNA methylation. Virions of type 2 CPV will also synthesise and methylate ssRNA transcripts of all 10 viral RNA genome segments in vitro. However, with this virus tvpe there is onlv a partial dependence on the presence of AdoMet, or structurally related compounds, for efficient ssRNA synthesis (Mertens and Payne, 1978). Recent studies have shown that the stimulatory effect of AdoMet is governed by some interaction between it and the virion enzymes which leads to an allosteric modification and results in an increase in
INTRODUCTION
S-Adenosyl methionine (AdoMet) acts as a methyl group donor during the in vitro synthesis of “capped” messenger RNA (mRNA) by the virion-associated polymerase of cytoplasmic polyhedrosis viruses (CPVs) (Furuichi, 1974, 1978a, 1981; Mertens and Payne, 1978). In addition, AdoMet massively stimulates RNA transcription by the polymerase of type 1 CPV. Other compounds related in structure to AdoMet can
’ Present address: Department of Biochemistry Animal Virus Research Institute, Pirbright, Woking, Surrey GU24 ONF’, United Kingdom. Author to whom requests for reprints should be addressed. * Present address: Department of Entomology and Insect Pathology, Glasshouse Crops Research Institute, Worthing Road, Littlehampton, West Sussex BN16 3P1J, United Kingdom. 0042-6822/83 fLWCI Copyriyht All rights
G 19E!3 hy Academic Prws, Inc. of reproduction in any form reserved.
18
CPV mRNA
SYNTHESIS
AND
polymerase activity (Furuichi, 1981). It has been suggested that AdoMet or its analogues bind to the active site of the methyl transferase(s) in the virions. In so doing, they might not only serve as methyl donors or competitive inhibitors of methylation but also stimulate RNA synthesis (Wertheimer d al, 1980). In this paper we present results which are inconsistent with this interpretation. We suggest that the effects of AdoMet or its analogues are due to their binding to at least two functionally distinct sites on the virus particle; the transcription control site and the methyl transfer site(s). In the present study, several compounds structurally related to AdoMet were examined for their effects on the polymerase activities of CPV types 1 and 2. The main compounds used were AdoMet, the D- and L-stereoisomers of AdoHcy (D-AdoHcy and r..-AdoHcy, respectively), S-adenosylL-ethionine (AdoEth) and adenosine. An analysis was also made of the ability of viral mRNA synthesized in vitro (in the presence of each of these compounds) to stimulate protein synthesis in a wheatgerm translation system. The results supported the conclusion that effective protein synthesis in vitro is dependent on the presence of complete methylated “cap” structures at the 5’ terminus of CPV mRNA molecules. MATERIALS
AND
METHODS
Viruses. The preparation of polyhedra and virus particles of CPV types 1 and 2 for these experiments, was identical to the methods described in a previous publication (Mertens and Payne, 1978). Electrophoretic analysis of RNA. Singlestranded (ss)RNA synthesised in the polymerase assay system was analyzed by electrophoresis using a 1.8% polyacrylamide, 0.6% agarose, 6 M urea gel system (Schuerch et &, 1975). After electrophoresis, the gels were stained with 0.01% ethidium bromide in 0.02 M acetate buffer (pH 7.8) and the ssRNA bands were visualised and photographed under ultraviolet illumination. Some gels containing “dual-labelled” ssRNA were sliced into l-mm slices, which were then treated with 200 ~1 of 60% (w/v) hydrogen peroxide overnight at 60”.
TRANSLATION
IN VZTRO
19
The dissolved slices were resuspended in 10 ml of a scintillation cocktail containing 66.7% toluene, 33.3% Triton X-100, 0.4% (w/v) 2,5-diphenyl-oxazole (PPO), and 0.05% (w/v) 1,4-di-2-(5-phenyl-oxazole) benzene (POPOP). The amount of “H and “P in each slice was determined in a Packard-Tricarb liquid scintillation spectrometer. Polymerase assays, Polymerase assay conditions were as previously described (Mertens and Payne, 1978). Structural analogues of AdoMet were included at concentrations indicated in the text. All analogues were obtained from the Sigma Chemical Co., Ltd. For the production of unlabelled ssRNA the standard polymerase assay was scaled up X30; radioactive UTP was omitted from the reaction system and the “cold” UTP concentration was increased to 2.0 mM. After incubation, virus particles and bentonite were removed by centrifugation (70,000 g for 1 hr at 4”), the supernatant was extracted with an equal volume of water-saturated phenol to remove any traces of ribonuclease, and then washed three times with diethyl ether. The ssRNA was precipitated by the addition of 0.25 vol of 10.0 M lithium chloride (sterile, final concentration, 2.0 M), followed by storage for 16 hr at 0” (Furuichi, 1974). The precipitate was pelleted by centrifugation (30,000 g, 10 min at 4“), washed twice with ethanol and once with diethyl ether, dried, resuspended in a small volume of sterile deionized water, and then stored frozen at -20”. The concentrations of these preparations of ssRNA were determined from their absorbance at 260 nm. Cell free translation assays. A cell free translation system was produced from wheat germ using the experimental procedure described by Roberts and Paterson (1973). The wheat-germ extract was stored as lOO-j.J aliquots in sterile glass vials under liquid nitrogen and transfer (t)RNA was not added to this assay system. The translation assays had a final volume of 25 ~1, containing 10 ~1 of wheat-germ extract, 20 mMHEPES buffer (pH 7.6), 2 mM dithiothreitol, 1 mM ATP, 20 pM GTP, 8 mM creatin phosphate, 40 pg/ml of creatin phosphokinase, 20-30 pM of the appropriate amino acids, 80 mM potassium chloride,
20
MERTENS
AND
2.5 mM magnesium acetate, 5 PCi of [3H]leucine, and CPV ssRNA as indicated in the text. Reaction mixtures were incubated for 1 hr at 25”. The reaction was stopped by the addition of 200 ~1 of 0.1 M KOH, followed by incubation at 37“ (to hydrolyse amino-acylated tRNA molecules). The level of protein synthesis was determined by measuring the amount of label incorporated into material precipitated by the addition of 10 ml of 10% TCA. After storage for 1 hr at 4”, the precipitate was collected on glass fiber filters, washed with 40 ml of 10% TCA, 10 ml of ethanol, and then dried. The filters were placed in 5 ml of a toluene-based scintillation cocktail containing 0.4% (w/v) PPO, 0.05% (w/v) POPOP, and then counted in a PackardTricarb liquid scintillation spectrometer. Translation assays were performed in duplicate or triplicate for each set of reaction conditions (as indicated in the text) and the average result for amino acid incorporation was calculated. RESULTS
Polymer-me Activity AdoMet, L-AdoHcy, D-AdoHcy, and AdoEth all stimulate the polymerase activity of CPV types 1 (Fig. la) and 2 (Table
PAYNE
1, Fig. lb). Of the nucleosides, nucleoside monophosphates and nucleoside diphosphates tested only adenosine increased ssRNA synthesis (Table 1). There was no observable stimulation of transcription caused by either homocysteine or methionine in this system (Table 1, Fig. 1). The addition of increasing concentrations of the stimulating compounds to reaction mixtures containing virions of either CPV type increased ssRNA synthesis up to a maximum level (Fig. 1). Beyond this level little or no increase of transcription was caused by the further addition of any of the other stimulation compounds tested (Table Z), suggesting that these compounds have a similar mode of action. When analysed at low concentrations, the most effective stimulator for either CPV was L-AdoHcy and the least effective was adenosine (Fig. 1). RNA Analysis Single-stranded RNAs synthesised in the presence of 0.5 mM AdoMet, L-AdoHcy, DAdoHcy, AdoEth, or 1.5 mM adenosine, were analysed by gel electrophoresis. The relative amounts of RNA synthesised (measured by UMP incorporation) in the presence of these compounds are shown in
c mMolar
concentration
of adenosine
analogue
FIG. I. A comparison of the effect on polymerase activity produced by the addition concentrations of AdoMet (0), I.-AdoHcy (O), AdoEth (Cl), D-AdoHey (A), Adenosine mocysteine (A) to assays containing virions of (A) type 1 and (B) type 2 CPV.
of varying (W, or ho-
CPV mRNA
SYNTHESIS
AND TABLE
TRANSLATION
21
IN VITRO
1
EFFECT OF A RANGE OF CHEMICALS ON CPV POLYMERASE Acme Nanomoies Addition to standard assay0
Molarity (mW
No addition AdoMet AdoMetb AdoMet L-AdoHcy * D- AdoHcy b AdoEthb
of UMP incorporated
Type 1 CPV (40 &ml)
Type 2 CPV (40 mz/ml)
0.05 0.5 1.5 0.5 0.5 0.5
0.33 7.25 7.36 8.81 8.10 8.25 8.30
1.80 4.21 5.27 4.01 5.40 5.51 5.60
0.13 0.10 0.13 0.12 0.11 0.10 0.11
AMP CMP GMP UMP
0.5 0.5 0.5 0.5
0.34 0.33 0.36 0.34
1.80 1.61 1.50 1.73
0.14 0.10 0.11 0.13
ADP CDP GDP UDP
0.5 0.5 0.5 0.5
0.35 0.33 0.32 0.34
1.64 1.60 1.71 1.75
0.11 0.11 0.13 0.14
Adenosine Adenosineb Cytidine Guanosine IJridine
0.5 1.5 0.5 0.5 0.5
0.93 3.87 0.34 0.34 0.35
3.26 4.20 1.62 1.81 1.73
0.13 0.15 0.11 0.10 0.14
Methionine Homocysteine
0.5 0.5
0.33 0.35
1.54 1.72
0.10 0.10
a Assays were incubated for 5 hr at 31” and were performed in triplicate. bssRNA synthesised in large-scale assays under these concentrations of stimulators elcctrophoresis (Fig. 2).
Table 1. The electrophoretic profiles of the products from the two viruses remained practically constant regardless both of the nature of the stimulator used (Fig. 2), or of different concentrations of AdoMet (0.05-1.5 mkf; results not shown). These experiments clearly demonstrate that although the specific nature and concentration of the stimulator used affect the overall level of transcription by these CPVs they have little effect on the relative levels of transcription of the different genome segments. Each RNA segment of both viruses is transcribed in approximately equal amounts by weight (Smith and Furuichi, 1980; Payne and Mertens, 1983).
Control (no virus)
were analyzed
by
RNA Methyl&cm A modified (dual label) polymerase assay system, containing [w~P]UTP to measure RNA synthesis and [methyl-3H]AdoMet to measure methylation, was used to analyse the effects caused by L-AdoHcy, D-AdoHcy, AdoEth, or adenosine on methylation of the polymerase products of CPV types 1 and 2. Using this system, it was found, as previously reported (Mertens and Payne, 1978; Furuichi, 1978a), that AdoHcy was an effective inhibitor of the methyl transferase activity of both CPV types (Figs. 3a, b, e, f). AdoEth also inhibited transmethylation (Figs. 3c, g) and appeared to be
22
MERTENS AND PAYNE TABLE2 COMPARISON OF THE STIMULATION OF CPV POLYMERASE ACWITY BY AdoMet, D-AdoHcy, AdoEth, AND ADENOSINE, ADDED SEPARATELY OR TIXETHER Nanomoles
Addition
to standard
No addition AdoMet, 0.5 mM L-AdoHcy, 0.5 m&f AdoEth, 0.5 mM D-AdoHcy, 0.5 mM Adenosine, 1.5 mM AdoMet, 0.5 mM t AdoMet, 0.5 m&f + AdoMet, 0.5 m&f t AdoMet, 0.5 m&f t L-AdoHcy, 0.5 mM L-AdoHcy, 0.5 mM L-AdoHcy, 0.5 mM D-AdoHcy, 0.5 mM D-AdoHcy, 0.5 mM AdoEth, 0.5 m&f + ‘Assays
were incubated
assay system’
LXst~tim
An analysis was made of the methyl distribution between the different size classes of the polymerase products separated by electrophoresis (Fig. 4). Although the definition achieved by the slicing of these gels was not sufficiently good to permit an estimate of the levels of methylation of each of the product species;
group
Type 2 CPV (40 &ml)
0.5 a.4 8.7 9.6 9.2 3.8 9.4 9.5 9.0 8.9 9.2 9.1 9.2 9.8 9.3 9.4
2.4 4.4 4.4 4.7 4.8 4.2 4.3 3.7 4.7 3.8 4.0 4.4 4.1 4.8 4.7 4.8
for 5 hr at 32.0” and were performed
of Methyl Groups
of UMP incorporated
Type 1 CPV (40 &ml)
L-AdoHcy, 0.5 mM AdoEth, 0.5 mA4 D-AdoHcy, 0.5 mM adenosine 1.5 mM t AdoEth, 0.5 mM t D-AdoHcy, 0.5 mM + adenosine, 1.5 mM + AdoEth, 0.5 mM + adenosine, 1.5 mM adenosine, 1.5 mM
slightly more effective in this respect (at lower concentrations) than I.-AdoHcy (Figs. 3b, f). The addition of AdoEth or LAdoHcy did not, however, cause any observable inhibition of methyl group incorporation until they were added to these assays at concentrations of approximately two to four times that of the [methyl‘H]AdoMet already present. In contrast, D-AdoHcy (a more effective stimulator of ssRNA synthesis than AdoMet) and adenosine (the least effective stimulator) did not greatly reduce methylation of the polymerase products of either type 1 (Figs. 3a, d) or type 2 (Figs. 3e, h) CPV at the concentrations tested.
L-AdoHcy,
in triplicate.
it is clear that, as with reovirus ssRNA (Levin and Samuel, 1977), the incorporated methyl groups were not restricted to the transcripts of any particular genome segment(s) but appeared to be distributed between the different size classes separated by this technique.
Translaticm of ssRNA The nature of the cap structure present at the 5’ termini of the ssRNA synthesised by these CPV in vitro, is affected by the nature of the stimulators present in the polymerase assay system (Furuichi, 1978a, b; Mertens and Payne, 1978, Wertheimer et al, 1980). The cap structures and the degree of methylation may well, in turn, affect the stability and messenger activity of these ssRNA molecules (Furuichi and Miura, 1975; Muthukrishnan et al, 1975; Shih et ah, 1976; Levin and Samuel, 1977; Shimotohno et aL, 1977, Furuichi, 1978a; Banerjee, 1980). Single-stranded RNA synthesised by CPV types 1 and 2 in the presence of each of the five stimulators tested (Fig. 2) was
CPV mRNA
SYNTHESIS
AND
TRANSLATION
IN
VITRO
Type 1
FIG. 2. Electrophoretic analysis of ssRNA synthesised in large-scale polymerase assays by CPV types 1 and 2 in the presence of 0.5 mML-AdoHcy (a), 0.5 mM AdoMet (b), 0.5 ~MD-AdoHcy (c), 0.5 mM AdoEth (d), or 1.5 mM adenosine (e). The same amounts of RNA were analysed on each gel despite differences in the levels of polymeraae activity under different reaction conditions (Table 1). The gels were photographed after staining with ethidium bromide.
adcled at varying concentrations to wheatassays. The messenger gel mm translation act ivity of each preparation of RNA was asslessed from its ability to increase amino aci d incorporation in this system (Fig. 5).
The preparations of “unmethylated” ssRNA synthesized by both viruses in the presence of L-AdoHcy, D-AdoHcy, or adenosine all produced very similar “stimulation curves” when tested in the wheat-germ system.
24
MERTENS Type
AND
PAYNE
1 CPV
Type 2 CPV
e 0 .
/@-H-O D-Ado
Hey
L b
f
a-0
\
L-Ado
HCY
l -.-+-*-------. Ado Elh
Adenosine
81
O-04
r-•-*-
g
d
-a
u
OLI
I
J
I
0
0.5
1.0
0
.
l -H-’
e-
Adenosine I 0.5
I 1.0
FIG. 3. A comparison of the effects of varying concentrations of (a and e) D-AdoHcy, (b and f) L-AdoHcy; (c and g). AdoEth; (d and h) adenosine on methyl group (0) and UMP (0) incorporation by virions of type 1 CPV (a-d) and type2 CPV (e-h) under standard “dual-label” reaction conditions, using [cI-~P]UTP and [%iJAdoMet.
4,300
6,500
A
750
900 r r*
.
VW7
600
0 0
80
0
80
Gel - length (mm) FIG. 4. Single-stranded RNAs of (A) type 1 and (B) type 2 CPV, dual-labclled using [(u-~P]UTP and [‘H]AdoMet, were analysed by electrophoresis. After staining, the gels were cut into l.O-mm slices and the amounts of =P and ‘H in each slice were determined. The position of the bands which were visible after staining are marked by arrows.
CPV mRNA
SYNTHESIS
AND
TRANSLATION
Zhr VZTRO
A
pg of RNA added per assay FIG. 5. The messenger activity of CPV ssRNA, synthesised in the presence of 0.5 mM AdoMet (O), 0.5 mM I.-AdoHcy (O), 0.5 mM D-AdoHcy (A), 0.5 mM AdoEth (Cl), and 1.5 mM adcnosine (m) in wheat-germ translation assays. The preparations of RNA used were the same as those analysed in Fig. 2. (A) Type 1 CPV RNA; (B) type 2 CPV RNA.
“Methylated” ssRNA, synthesised in the presence of AdoMet increased amino acid incorporation considerably. This difference was particularly noticeable with the preparations of ssRNA from type 2 CPV (Fig. 5b), possibly because the average level of methylation in the products from the type 2 CPV polymerase is higher than in those from type 1 CPV (see Discussion). RNA synthesised by either CPV type in the presence of AdoEth, caused a lower level of stimulation than the ssRNA synthesised in the presence of any of the other four transcription stimulators tested. Some capping and methylation of ssRNA may occur in the wheat-germ assay system (Levin and Samuel, 1977). It was therefore possible that the differences in messenger activity of the different preparations of CPV ssRNA were at least partially masked under standard translation assay conditions. In view of this possibility, an analysis was made of the effects produced by the addition of AdoMet, L-AdoHcy (a stimulator and inhibitor of transmethylation, respectively), and the other CPV transcription stimulators (D-AdoHcy, AdoEth, and adenosine) to the wheat-germ assay system. The results obtained using “un-
methylated” RNA synthesised in the presence of D-AdoHcy and adenosine were similar to those obtained with L-AdoHcy and are not shown in Fig. 6. The addition of AdoMet to the translation assays increased amino acid incorporation in response to “unmethylated” ssRNA of either CPV type (synthesised in the presence of L-AdoHcy; Figs. 6b, e). Some increase in amino acid incorporation was also observed when AdoMet was added to assays containing “methylated” ssRNA (synthesised in the presence of AdoMet; Figs. 6a, d). However, the overall increase, particularly with ssRNA from type 2 CPV (which has a higher level of methylation) was considerably less than with unmethylated material (compare Fig. 6d with Fig. 6e). When AdoMet was added to translation assays containing ssRNA synthesised in the presence of AdoEth, there was only a small increase in amino acid incorporation comparable to that produced in the presence of methylated ssRNA (Figs. 6c, f). This may well be due to the incorporation of ethyl groups into the RNA synthesised in the presence of AdoEth (Furuichi, 1978a, b), preventing their subsequent methyl-
26
MERTENS Type
AND
PAYNE
1 CPV
Type 2 CPV a
0
0.5
1.0
0
0.5
m~olarconcentrationof adenosineanalogue FIG. 6. The effects of the addition of AdoMet (0), r.-AdoHcy (O), D-AdoHcy (A), AdoEth (O), or adenosine (W) to wheat-germ translation assays containing (a and d) “mcthylated” CPV ssRNA (synthesised in the presence of AdoMet), (b and e) “unmethylated” ssRNA (synthesised in the presence of L-AdoHcy); and (c and f) “ethylated” ssRNA (synthesised in the presence of AdoEth).
ation. None of the other compounds tested (L-AdoHcy, AdoEth, D-AdoHcy, or adenosine) increased amino acid incorporation when added to the wheat-germ system. DAdoHcy appeared to have little effect on the level of protein synthesis, regardless of the ssRNA preparation used. Adenosine inhibited protein synthesis and at the highest concentration tested (1.0 mM), reduced amino acid incorporation to near background (endogenous activity). The remaining compounds, L-AdoHcy and AdoEth which both inhibited transmethylation reactions in the polymerase assay system inhibited protein synthesis to an almost identical extent when added to the wheat-germ assay system. This effect, like the stimulation caused by AdoMet was most pronounced in assays containing “unmethylated” CPV ssRNA (Figs. 6b, e).
DISCUSSION
The original discovery that the methyl donor, AdoMet, increased ssRNA synthesis by type 1 CPV polymerase first suggested that transcription and methylation of the RNA were linked (Furuichi, 1974). That this was not the case was clearly demonstrated by further studies using analogues of AdoMet which are not methyl donors but which, nonetheless, increase polymerase activity (Mertens and Payne, 1978; Furuichi, 19’78a;Wertheimer et aL, 1980). In the present study, we have compared the effects of AdoMet and structurally related compounds on the transcription and translation of CPV ssRNAs. Similar relative levels of the ssRNA transcripts of all the viral genome segments were produced regardless of the different molecular
CPV mRNA SYNTHESIS AND TRANSLATION IN VITRO structures of the compounds which stimulated polymerase activity. In addition, all these compounds appeared to increase transcription by a basically similar mechanism as the “maximum” level of polymerase activity could not be increased when AdoMet and its analogues were added in combination, rather than singly. Perhaps of greatest interest were the differential effects of these compounds on RNA methylation; in particular, the different responses of the viruses to the stereoisomers, D- and L-AdoHcy. Both compounds increased polymerase activity but, while L-AdoHcy inhibited methylation, DAdoHcy did not greatly reduce methyl group incorporation into the RNA transcripts. Previous studies have suggested that the most likely explanation for the effect of compounds such as AdoMet or AdoHcy, is that they bind to a single site on the virus and both stimulate RNA synthesis and serve as methyl donors or competitive inhibitors of methylation (Wertheimer et a,l., 1980). D-AdoHcy does not fit into this pattern. Nor do other compounds which inhibit CPV-mRNA synthesis in vitro without reducing methylation. These include compounds such as S-guanosyl-L-methionine(GuaMet),S-uridyl-L-methionine(UriMet), and AdoHcy-dialdehyde (Wertheimer et cd., 1980). However, these anomalies can be interpreted by an alternative model (also considered by Wertheimer et al, 1980) in which molecules of AdoMet or its analogues bind separately to functionally distinct sites, i.e., one to a methyl transfer site(s) (Payne and Mertens, 1983), the other to a transcription control site. While AdoMet and L-AdoHcy may bind efficiently to both sites, DAdoHcy, GuaMet, UriMet, and AdoHcydialdehyde may only bind to the site involved in transcription control. In the latter group of compounds, only the binding of D-AdoHcy leads to increased transcription. In contrast to type 1 CPV, the transcription control site in type 2 CPV appears to be under less rigid control by AdoMet or its analogues, as the virion polymerase is relatively active in their absence. The different responses of the polymerase of both CPV types 1 and 2 to AdoMet
27
and its analogues are useful in evaluating the structural features of the molecules which are important in the promotion of efficient transcription. Although adenosine increased polymerase activity, the most effective compounds tested by us, bore an amino acid side chain. Differences in the nature and conformation of the side chain (including the sulphonium group) gave minor differences in polymerase activity, although the amino acids homocysteine or methionine by themselves did not stimulate transcription. These results are in general agreement with those of Wertheimer et al. (1980), and suggest that no single structural feature has yet been defined as the key component in CPV polymerase stimulation. However, it is interesting that in a recent study, Wu et nl. (1981) have shown that methyl-methionine alone is also very effective in increasing polymerase activity in type 1 CPV virions. Studies on the methylation of CPV-RNA showed, as with reovirus (Levin and Samuel, 19’7’7) that methyl groups were not restricted to any specific genome segments. There does not, therefore, appear to be any control of the relative levels of expression of the different genome segments of CPV types 1 and 2, mediated either by the differential stimulation of transcription of a particular segment, or by an uneven distribution of methyl groups between transcripts. Assuming that the majority of the radioactive substrates incorporated into TCA precipitable material in the dual-label experiments represented macromolecular components of newly synthesised RNA and does not include the small oligonucleotides produced in the assay (Smith and Furuichi, 1982), it is possible to calculate the average number of methyl groups attached to each ssRNA chain. Taking into account the relative frequency of transcription of the different size classes of RNA (Payne and Mertens, 1983), the average MW of the ssRNA chains was 0.42 X 10’ and 0.52 X 10” for the products of CPV types 1 and 2, respectively. Taking the average uridine content as 28.5 and 31.5% for the RNA of CPV types 1 and 2, respectively (Miura et al, 1968; Payne and Churchill, 1977), it was calculated that type 2 CPV incorporated
28
MERTENS
an average of 1.0 methyl group per RNA molecule and type 1 CPV incorporated 0.52 groups per RNA molecule, under standard dual-label reaction conditions. These values are considerably lower than those published by us from similar experiments as it was incorrectly assumed that the segments of the CPV genome were all transcribed with approximately equal frequency (Mertens and Payne, 1978) rather than in equivalent amounts by weight (Smith and Furuichi, 1980; Payne and Mertens, 1983). From reports by other workers on the 5’ terminal structure of type 1 CPV RNA, it is probable that methyl groups are incorporated into a blocked or cap structure (m7GpppAmpGp . . .) at the 5’ termini of the polymerase assay products (Furuichi and Miura, 1975; Furuichi, 1978a; Shimotohno and Miura, 1974, 1976). The results of the analysis of the messenger activity of CPV ssRNA in the wheat-germ assay system are consistent with a requirement for methylation of the 5’ terminal cap structure for efficient initiation of translation (Levin and Samuel, 1977; Shimotohno et aL, 1977; Banerjee, 1980). The preparation of ssRNA which caused the highest specific stimulation of protein synthesis was that synthesised by type 2 CPV in the presence of AdoMet (“methylated” ssRNA). The ssRNA from type 1 CPV synthesised under similar reaction conditions caused a smaller increase in amino acid incorporation, possibly as a result of the lower level of methylation observed in this material. The “unmethylated” ssRNA preparations from both CPVs (synthesised in the presence of D-AdoHcy, L-AdoHcy, or adenosine) produced very similar stimulation curves, suggesting that in the “unmethylated” form (Furuichi, 1981), there is little intrinsic difference in the translational efficiency of ssRNA from these two viruses. The ssRNA synthesised by either CPV type in the presence of AdoEth produced the lowest levels of stimulation of those preparations tested. This may well result from the incorporation of ethyl groups at the 5’ terminal cap structure (Furuichi, 1978b) and suggests that ethylated cap structures block efficient translation in this system.
AND
PAYNE
In conclusion, the polymeraseand methylation-control mechanisms within the virions of CPVs do not appear to affect the relative levels of expression of the different CPV genome segments. It seems probable that these control mechanisms will help to ensure a high level of capping and methylation of all of the ssRNA products. However, if the end products of the transmethylase and polymerase reactions (AdoHcy and pyrophosphate, respectively) can accumulate during CPV replication, then it is possible that some progressive reduction could occur in the levels of capping and methylation at later stages of infection. Such changes could have considerable significance in affecting the processing and stability of the viral messenger RNA (Furuichi and Miura, 1975; Banerjee, 1980; Payne and Mertens, 1983). ACKNOWLEDGMENTS The authors would like to thank Dr N. F. Moore for his assistance in setting up the wheat-germ system and Dr K. A. Harrap for useful discussion of experimental procedures. REFERENCES BANEKJEE, A. K. (1980). 5’-terminal cap structure in eukaryotic messenger ribonucleic acids. Mirrohiol. Rev. 44, 175-205. FUKUI~HI, Y. (1974). Methylation-coupled transcription by virus associated transcriptase of cytoplasmic polyhedrosis virus containing double-stranded RNA. Nuu?eic Aa& Res. 1,809-822. FUKIJI~HI, Y. (1978a). Pre-transcriptional capping in the biosynt.hesis of cytoplasmic polyhedrosis virus mRNA. Proc Not. Acud Sci USA 75,1086-1990. FURUICIII, Y. (19’78b). Stimulation of cytoplasmic polyhedrosis virus mRNA synthesis by S-adenosylmethionine and its derivatives. “Abstracts of the Fourth International Congress for Virology, The Hague,” p. 332. Ccntre for Agricultural Publishing and Documentation, Wageningen. FURUICHI, Y. (1981). Allosteric stimulatory effect of S-adenosyl methionine on the RNA polymeraae in cytoplasmic polyhedrosis virus. A model for the positive control of eukaryotic transcription. J. BioL Chem 256, 483-493. FURUICHI, Y., and M~JRA, K. (1975). A blocked structure at the 5’ terminus of mRNA from cytoplasmic polyhedrosis virus. Nuture (Lmdm) 26.3, 374-375. LEV~N, K., and SAMUEL, C. E. (1977). Biosynthesis of reovirus specified polypeptides. Effect of methylation on efficiency of rcovirus genome expression in vitro. Virokgy 77, 245-259.
CPV mRNA
SYNTHESIS
AND
MERTENS, P. P. C., and PAYNE, C. C. (1978). S-adenosylL-homocysteine as a stimulator of viral RNA synthesis by two distinct cytoplasmic polyhedrosis viruses. .I. ViroL 26, 332-835. MIURA, K., FUJII, I., SAKAIU, T., FUKE, M., and KAWASE, S. (1968). Double-stranded ribonucleic acid from cytoplasmic polyhedrosis virus of the silkworm. J. ViroL 2, 1211-1222. MUTHIJKRISHNAX, S., BOTH, G. W., FURUICHI, Y., and SHATRIN, A. J. (1975). 5’ terminal ‘I-methyl guanosine in eukaryotic mRNA, required for translation. Nature &m&m) 255,3337. PAYNE, C. C., and CHURCHILL, M. P. (1977). The specificity of antibodies to dsRNA in antisera to three distinct cytoplasmic polyhedrosis viruses. Virology, 79, 251-258. PAYXE, C. C., and MERTENS, P. P. C. (1983). Cytoplasmic polyhedrosis viridae. In “The Reoviridae” (W. K. Joklik, ed.), pp. 425-504, Plenum, New York. ROBERTS, B. E., and PATERSON, B. M. (1973). Efficient translation of tobacco mosaic virus RNA and rabbit globin ts RNA in a ceil-free system from commercial wheat germ. Proc. Nat. AU& sci, USA 70, 2330‘2334. SCHUERCH, A. R., MIT~HGLL, W. R., and JOKLIK, W. K. (1975). Isolation of intact individual species of single and double-stranded RNA after fractionation by polyaerylamide gel electrophoresis. Anal. Biochem 65, 331-345. SIIIH, D. S., DASGUPTA, A. R., and KAESBERG, P. (1976). ‘I-methyl-guanosine and the efficiency of RNA translation. J. ViroL 19. 637-642
TRANSLATION
IN
VITRO
29
SHIMOTOHXO, IL, and Mums, K.-I. (1974). <5’-terminal structure of messenger RNA transcribed by the RNA polymerase of silkworm cytoplasmic polyhedrosis virus containing double-stranded RNA. J. MoL &id 86.21-30. SHIMOTOHNO, K., and MIVRA, K.-I. (1976). The process of formation of the S’terminai modified structure in messenger RNA of cytoplasmic polyhedrosis virus. FEBS Z&t 64.204-208. SHIYOTOHNO, K., KODAMA, Y., ~~~~~~~~~~~~ J., and MICRA, K.-I. (1977). Importance of g-terminal blocking structure to stabilize mRNA in cukaryotic protein synthesis. E’roc. Nat Acd Sci USA 74, 2734-2738. SMITH, R. E., and FUR~JI~I-II, Y. (1980). Gene mapping of cytoplasmic polyhcdrosis virus of silkworm by the full-length mRNA prepared under optimized conditions of transcription in vitro. Virology 103, 279-290. WERTHEIMER, A. M., CHEN, S. Y., BORCHARI)T, R. T., and FIJRUICHI, Y. (1980). S-adenosyl methionine and its analogs: Structural features correlated with synthesis and methylation of mRNAs of cytoplasmic polyhedrosis virus. J. BioL C&-m. 255,59245930. WV, A., DAI, R., SHEN, X., and SUN, Y. (1981). [‘IImethyl]-methionine as possible methyl donor for formation of ti’-terminus of in vitro synthesized mRNA of cytoplasmic polyhedrosis virus of silkworm, Bwntn~;c mot-i. Ski Sin 24,1737-1742.