- Email: [email protected]
The effect of poliovirus infection on the translation In Vitro of VSV messenger ribonucleoprotein particles
VIROLOGY
129.415-430
(1983)
The Effect of Poliovirus Infection on the Translation In Vitro of VSV Messenger Ribonucleoprotein Particles CHARLOTTE L. JONES
AND
ELLIE EHRENFELD’
Departments of Cellular, Viral, and Molecular Biology and Biochemistrg University of Utah School of Medicine, Salt L.&e City, Utah 85152 Received April
7, 1985;accepted June 20, 1989
Messenger ribonucleoprotein (mRNP) particles were isolated and examined for the presence of factors involved in the inhibition of protein synthesis induced by poliovirus infection. Vesicular stomatitis virus (VSV) mRNPs were used as a model for cellular mRNPs. These mRNPs were translated in HeLa cell extracts with a similar efficiency and optimal conditions to that of purified mRNA, but they were not translated in extracts prepared from poliovirus-infected HeLa cells, which have been shown to be defective in cap-binding protein activity. We conclude that mRNP proteins do not include cap-binding protein activity, since the mRNPs were not able to bypass the restriction on translation of capped mRNAs in polio-infected cell extracts. In addition, VSV mRNPs were isolated from polio-superinfected cells, in which their translation was inhibited. These mRNPs were translated in vitro as well as normal VSV mRNPs. No evidence of a modification or a blocking factor on the mRNPs which prevented their translation following polio infection was observed. Thus, within the limits of the in vitro translation assays used, no factors involved in the discrimination between polioviral and cellular or VSV mRNA could be detected in the mRNP particle. INTRODUCTION
Poliovirus infection of HeLa cells results in an inhibition of host cell protein synthesis demonstrated to occur at the initiation step of translation (Ehrenfeld and Manis, 1978; Kaufman et al, 1976; Liebowitz and Penman, 1971). Extracts from poliovirus-infected cells are unable to translate capped mRNAs in vitro (Rose et aL, 1978; Sonenberg et aL, 1979), and several investigators have reported alterations in the properties of cap-binding protein complexes isolated from infected cells (Etchison et aL, 1982; Hansen et aL, 1982a). It is likely that a virus-induced defect in the function of the cap-recognition system is responsible for the failure of cellular mRNA to bind to ribosomes and initiate translation (Blobel, 1973). ’ Present address: Department of Cellular and Developmental Biology, Biosciences West Building, University of Arizona, Tucson, Arizona 85721. ‘To whom reprint requests should be addressed. 415
The population of host cell mRNA present in infected cells has been shown to remain structurally intact and fully functional for translation in vitro (Ehrenfeld and Lund, 1977; Fernandez-Munoz and Darnell, 1976), despite its failure to be utilized in the infected cell. However, it has been shown that eukaryotic mRNA is associated in the cell with proteins to form a messenger ribonucleoprotein particle (mRNP) (Greenberg, 1980; Spirin, 1969). Numerous studies have been conducted in an attempt to understand the structural and functional role of the proteins in mRNP particles on translation of mRNA, either as inhibitory or blocking proteins which might prevent translation under specific physiologic conditions, or as structural regulators which may affect the secondary structure of mRNA so as to expose or facilitate the function of ribosomebinding sites. The purpose of this study was to determine whether any factors involved in the translational discrimination between polioviral and cellular mRNAs in 0042~6822183$3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
416
JONES
AND
the infected cell occur in the mRNP particle. A number of mRNP proteins have been described, but there is little consensus on either the number or molecular weights of these proteins, which vary with the method of isolation and the system being analyzed. Recently, ultraviolet light-induced crosslinking has been successfully utilized to identify proteins physically associated with mRNA (Greenberg, 1980). Two tightly bound proteins, 52K and 78K M,, have been identified in most systems examined (Blobel, 1972; Blobel, 1973; Greenberg, 1980). The latter protein is preferentially associated with the poly(A) sequence of the mRNA (Blobel, 1973; Greenberg, 1980), and cross-reacts with antibodies made against the poly(A) polymerase (Rose et al, 1979). Protein kinase activity has been found associated with mRNPs in a number of systems (e.g., Egly et a& 1976). Beyond that, little is known regarding the function of the mRNP proteins, but there are several situations which suggest that mRNA-associated proteins can exert both positive and negative regulatory effects. Although none of the characterized eukaryotic initiation factors have been identified among the tightly bound mRNP proteins (Barrieux and Rosenfeld, 1979), a number of initiation factors have been shown to bind mRNA in vitro. These include eIF2 (Hellerman and Shafritz, 1975), eIF4B (Shafritz et al, 1976), eIF5 (Shafritz, et al., 1976), and the cap-binding proteins (Sonenberg et al, 1978). Other activities found associated with mRNPs include a met-tRNAf binding protein (Barrieux and Rosenfeld, 1979; Hellerman and Shafritz, 1975; Kaempfer et al, 1978) and a factor which supports the binding of isolated mRNPs to the 40 S ribosomal subunit from HeLa cells (Liautard, 1979). In several differentiating or developmentally regulated systems, the proteins associated with mRNP particles have been shown to exert a negative regulation of translation. For example, mRNPs isolated from unfertilized sea urchin eggs, which are not translated in vivo, are also not translated in vitro. However, the mRNA extracted from these particles is translated
EHRENFELD
well in vitro. Both mRNA and mRNPs from fertilized eggs are translated efficiently (Jenkins et aZ., 1978). To examine the role of mRNP proteins in polio-infected cells, we have used vesicular stomatitis virus (VSV) mRNPs as a model for cellular mRNPs, since VSV protein synthesis is inhibited by poliovirus infection in the same manner as is host cell translation (Doyle and Holland, 1972; Ehrenfeld and Lund, 1977). In addition, the VSV mRNAs and polypeptides are readily identified by gel electrophoresis, and Grubman and Shafritz have previously characterized the VSV mRNP particles (Grubman and Shafritz, 1977). The studies described in this report address the question of whether functional mRNP particles contain the cap-recognition activity, bound to mRNA. Such an activity would allow mRNPs to bypass the restriction imposed on the translation of capped mRNAs in a polio-infected cell extract, which is depleted of a functional capbinding protein activity. We show that VSV mRNPs, although translationally active, are not translated in a polio-infected cell extract. In addition, VSV mRNPs isolated from a poliovirus superinfected cell contain no detectable alteration or blocking activity which prevents their translation in vitro. MATERIALS
AND
METHODS
Virus and cells. HeLa S-3 cells were grown in spinner culture in Eagle’s minimal essential medium (Joklik’s modification) supplemented with 10% heat-inactivated calf serum. Vesicular stomatitis virus (VSV) stocks were grown in HeLa cells and purified as follows: Approximately 3 X 10’ cells (5 X 106/ml) in medium without serum were infected at a multiplicity of infection of l3 PFU/cell, and the virus was adsorbed for 1 hr at 37”. At this time the cells were diluted to a concentration of 106/ml and the medium was made 2% serum. The infected cells were incubated 18 hr at 34” and the medium was then cleared of cells and cell debris by centrifugation at 800 g for 15 min. The cleared supernatant was
TRANSLATION
OF mRNPs
made 2.2% NaCl and 7% polyethylene glyco1 6000 and stirred on ice for 5-6 hr. The virus precipitate was collected by centrifugation at 1800 g for 10 min at 4” and resuspended in 6-8 ml ET buffer (10 mM Tris-Cl, 1 mM EDTA, pH ‘7.4), sonicated 2-4 min in a Melter ultrasonic water bath, and layered on six lo-50% (w/w) sucrose gradients in ET buffer. The gradients were spun for 13 hr at 26,000 rpm, 4”, in a Beckman SW 28 rotor and the visible virus bands were collected with a syringe. The virus was diluted 3-4 X with cold ET buffer and collected in pellets by centrifugation in the SW 28 rotor at 23,000 rpm for 2 hr at 4”. The virus pellets were resuspended in ET, briefly sonicated, and layered on three 5-20% sucrose (w/w in ET buffer) gradients. The gradients were spun in the SW 28 rotor at 18,000 rpm for 50 min at 4” and the virus collected, diluted, and pelleted as before. The virus pellets were resuspended in l-2 ml ET containing 10% DMSO, and aliquots were stored at -70”. Virus titers were obtained by counting plaques on monolayer HeLa cells in Eagle’s minimal essential medium supplemented with 2% heat-inactivated calf serum. Poliovirus (Mahoney strain) type 1 was grown in HeLa S-3 cells in spinner culture. Cells were infected at a concentration of 5 X lo6 cells/ml with a multiplicity of infection of 20 PFU/cell, and the virus was adsorbed for 30 min at 37”. After adsorption, serum was added to 2% and the culture was incubated at 37” for another 5% hr. The cells were then harvested by centrifugation, washed twice with Earle’s salts solution, and resuspended in RSB (10 mM Tris-Cl pH 7.4, 10 mM NaCl, 1.5 mM MgClJ. The cells were broken by freezethawing three times and NP-40 was added to 1%. Nuc1e.i were removed by 5 min centrifugation at 1300 g and the supernatant was cleared by centrifugation at 10,000 g for 15 min. This supernatant was then diluted with RSB and made 1% SDS for a final volume of 30 ml, and the virus was collected in a pellet by centrifugation in an angle 50.2 Ti rotor at 30,000 rpm for 3 hr at 20”. Pellets were resuspended by stirring on ice with a magnetic stir bar in 4.5 ml of 10 mM Tris-Cl, pH 7.4,lO mM NaCl,
AFTER
POLIO
INFECTION
417
1 mM EDTA, 1% Brij 58, and 2.25 g CsCl was added to make a final density of 1.33 g/ml. The virus was then banded by spinning in an SW 50.1 rotor at 40,000 rpm, 17 hr, 4”. The visible virus band was removed with a pipet and stored at 4”. The virus was quantitated by ultraviolet absorbance at a wavelength of 260 nm where 1 absorbance unit = 1.3 X 1013 particles/ml with a particle/PFU ratio of 100. Plaque assay and RNA synthesis rates confirmed the calculated titer. mRNP isolations. An outline of the procedure is shown in Fig. 1. HeLa cells (5 X lo* cells/100 ml) were infected with VSV at a multiplicity of lo-20 PFU/cell at 37”. After 1 hr serum was added to 2%. AT 2.5 hr, actinomycin D was added to 2 pg/ml and [3H]uridine (New England Nuclear, 42 Ci/mmol) was added to 25 &i/ml. At 5.5 hr postinfection, the cells were collected, washed three times with Earle’s salts solution, and resuspended in 3 ml RSB. NP40 was added to 0.5%; the nuclei were removed by centrifugation and washed with 1 ml RSB and 0.5% NP-40 and the wash was added to the cytoplasmic extract. The cytoplasmic extracts were loaded directly on a 7-47% (w/w) sucrose gradient in RSB, and spun in a Beckman SW 28 rotor at 16,000 rpm for 15 hr at 4” (35,000 g). The gradient was collected into 30 fractions and small samples of each fraction were analyzed for TCA-precipitable radioactivity. The appropriate fractions for polysomes and free mRNPs were pooled as indicated in Fig. 2A and diluted approximately twofold with RSB. Polysomes and mRNPs were pelleted by centrifugation in a Beckman angle 50 Ti rotor at 48,000 rpm (160,000 g) for 24 hr at 4”. The pellets were each suspended in 0.5 ml NEB (0.15 MNaCl, 10 mM Tris, 10 mM EDTA, pH 7.4) by stirring with a magnetic stir bar on ice for 2-3 hr. The samples were then each layered on 1530% sucrose (w/w) gradients in NEB and spun in the Beckman SW 28 rotor at 25,000 rpm for 14.5 hr at 4” (85,000 g). Gradients were collected as before. The mRNP-containing fractions were pooled as indicated (Figs. 2B, C). Each pooled sample was divided in half; one half was made 0.5 MKCl and diluted approximately twofold with
JONES
418
AND
EHRENFELD
CYTOPLASM
SUCROSE
VELOCITY
GRADIENT
‘FREE
POLY SOMES
t
PELLETED
PELLETED
SUCROSE
IN
POLY SOt44L
r
PELLETED
SUCROSE
VELOCITY
GRADIENT
WASHED
VELOCITY IN EDTA
FREE mRNP ‘S
I 1 t
WASHED
IN
0.5
M KC1 3 PELLETED
t
t
POLYSOMAL
POLYSOKAL
mRNP ’ S
GRADIENT
EDTA
mRNP ‘5
0.5
mRNP’S
PELLETED
FREE
mRNP’S
IN
M KC1
PELLETED
FREE mRNP’S
SW
mRNP ‘S SW
FIG. 1. Scheme for isolation
of VSV mRNPs. Details
HSB (0.5 M KCl, 3 mM MgClz, 50 m&f Tris, pH 7.4); the other half was simply diluted twofold with NEB. The mRNPs were pelleted as before, the pellets rinsed gently and resuspended as before in 0.1 ml 10 mM Tris, 10 m&f KCl, pH 7.4, and the samples stored frozen at -70”. All solutions and glassware were autoclaved and the samples were always kept cold Gel analysis ofRNA. RNA was extracted with phenol from mRNP preparations or from cytoplasmic extracts and precipitated with 2.5 vol of 95% ethanol overnight at -70”. RNA was pelleted in an Eppendorf airfuge for 5 min (150,000 9) and the pellets were dissolved in 25 ~1 HzO. To each sample was added 25 ~1 sample buffer (borate buffer, 20% glycerol, 0.005% bromophenol blue, 10 mM methyl mercury hydroxide) and incubated 15 min at room temperature. The samples were then loaded on a 1% agarose horizontal slab gel in borate buffer (0.05 M boric acid, 5 mM sodium borate, 10 mM sodium sulfate, 1 mM disodium EDTA) which was 5 mM methyl mercury hydroxide. The gel was run at 90 V with
are described
under Methods.
the borate buffer in buffer trays recirculated. The gel was soaked in 500 ml 10% acetic acid, 10 m&f cysteine for 15 min to complex the mercury. Two methanol washes of at least an hour each followed, and the methanol was pulled from the gel under vacuum for 15 min. The gel was soaked 2 hr in PPO in methanol (32 g PPO/ 250 ml methanol), rinsed three times with Hz0 for 5 min each, sealed in Saran Wrap, and exposed to film at -70”. Oligo(dT) cellulose chromutography. RNA samples were dissolved in 0.5 M KCl, 10 mMTris, pH 7.4, at a concentration of less than 0.5 mg/ml and fractionated over a 0.5-g oligo(dT) cellulose (Collaborative Research, Inc.) column equilibrated with 0.5 M KCl, 10 mlM Tris-HCl, pH 7.4. The column was then washed with 5 ml of the same buffer and the bound fraction was eluted with 10 mlM Tris-HCl, pH 7.4. RNA samples were then ethanol precipitated for further use. Translation systems. Two translation systems were used: an unfractionated HeLa SlO extract, prepared from either
5
FRACTION
IO
NUMBER
I5
20
25 FRACTION
NUMBER
TOP
5
I,,,,
l,f,,,!,!
FRACTION
IO
t
I5
I.*.-.
NUMBER
I
\
t
I,,,,, 20
25 TOP
and free mRNPs. Cytoplasmic extracts of VSV-infected to maintain polysome structure. Fractions indicated by The polysome sample from (A) was made 20 m&f EDTA as described under Methods. The horizontal bar indicates (A) was made 20 mM EDTA and sedimented as in panel centrifugation. The solid line indicates the absorbance at
BOTTOM
BOTTOM
TOP
,I,,, 0
30
3-
4-
o!
6-
65-
7-
7-
9-
io- c
e-
4
FIG. 2. Sucrose gradient fractionation for isolation of VSV mRNPs. (A) Separation of polysomal cells were sedimented through a sucrose gradient containing Mg+‘, as described under Methods, horizontal bars were pooled, and the mRNPs from each pool were collected by centrifugation. (B) to dissociate ribosomes and mRNPs, and was fractionated in a sucrose gradient containing EDTA, those fractions pooled to obtain the polysome-released mRNPs. (C) The free mRNP sample from B. The fractions indicated by the bar were pooled to obtain free mRNPs, which were collected by 260 nm; the closed circles show the [3H]uridine-labeled RNA in 0.1 ml of each 1.3-ml fraction.
BOTTOM
0
E ”
c) 0 8.
40s t
‘8
3
60s :
420
JONES
AND
uninfected or poliovirus-infected cells, and a fractionated HeLa cell system composed of salt-washed ribosomes, the ribosomal salt wash, and a soluble cell fraction. To prepare HeLa cell SlO extracts, cells were swollen in 2 cell vol of PSB (10 mM KCl, 2.5 mM DTT, 1.2 mM Mg acetate, 20 mM HEPES, pH 7.4) on ice for 10 min. Dounce homogenization (20 strokes) broke the cells, centrifugation at ‘700g for 4 min removed the nuclei, and centrifugation at 10,000g for 10 min cleared the cytoplasmic extract. This extract was then treated with micrococcal nuclease at a concentration of 20 pg/ml in 2 mM CaCla for 10 min at 18” according to the method of Pelham and Jackson (1976). The nuclease reaction was stopped by adding EGTA to 4 mM and the extract was immediately chilled and filtered through a Sephadex G-25 column (15 x 1.5 cm) equilibrated with PSB. Aliquots of the excluded material were stored frozen at -70’. Extracts from polio-infected cells were prepared in the same manner from cells which had been infected as described above and harvested 3’/2 hr postinfection. For the fractionated cell system, HeLa cell extracts were prepared and nuclease treated as described above; however, after the nuclease treatment, ribosomes were collected by centrifugation in a Beckman angle 50 Ti rotor at 48,000 rpm for 90 min at 4”. The supernatant (soluble cell fraction) was saved and concentrated (see below). The ribosomal pellets were then resuspended in PSB at a concentration of 250 AacO/ml on ice with a magnetic stir bar. To remove the initiation factors from the ribosomes, 2 M KC1 was added to 0.5 M and they were stirred 10 min on ice, after which the ribosomes were separated from the ribosomal salt wash by centrifugation in an angle 50 Ti rotor at 48,000 rpm, 4”, 90 min, and the ribosomal pellet was resuspended to give a concentration of 250 AzW/ml in 0.25 M sucrose, 0.2 mM EDTA, 1 mM DTT. Aliquots of the salt-washed ribosomes were stored at -70”. For a source of initiation factors, the ribosomal salt wash was dialyzed against 20 mMTris, pH 7.4,lOO mM KCl, 0.2 mM EDTA, 7 mM P-mercaptoethanol, and 5% glycerol, and aliquots were stored frozen at -70”.
EHRENFELD
The postribosomal supernatant was diluted in 2 vol of cold 10 mM P-mercaptoethanol, and 1 M acetic acid was added dropwise until the pH was 5.0. The resulting precipitate was collected by centrifugation at 4000 g for 5 min. The pellet was dissolved in 50 mM KCl, 3 mM Mg acetate, 30 mM Tris, pH 7.4, and 10 mM P--mercaptoethanol in l/10 the original volume. The pH was readjusted to 7.4 with KOH, and this fraction was stored frozen at -70”. Translation reactions (50 ~1) contained either 40% (v/v) SlO extract or 10% (v/v) salt-washed ribosomes, 2% (v/v) pH 5 soluble fraction, and 12% (v/v) ribosomal salt wash. In addition, the reactions contained 65 mM KCl, 28 mM HEPES, pH 7.4, 2.5 mM Mg acetate, 3 mM DTT, each amino acid except methionine at 4 PM, 2.5 pg tRNA, 1 mM ATP, 0.2 mM GTP, 25 mM creatine phosphate, 3 &i [35S]methionine (Amersham, 1220 Ci/mmol), and l-2 /*g mRNA. Reactions were incubated for 1 hr at 3O”C, diluted with SDS-PAGE sample buffer, and analyzed on 10% polyacrylamide gels (Laemmli, 1970). RESULTS
Isolation
of VSV rnRNPs
Following poliovirus infection, cellular mRNPs accumulate in the cytoplasm, not translated, and not associated with polysomes (Penman et ah, 1963). Poliovirus superinfection of VSV-infected cells produces a similar accumulation of untranslated VSV mRNPs (unpublished observations). To compare VSV mRNPs present in polioinfected cells with functional VSV mRNPs, it was first necessary to examine both the polysome-associated and the naturally occurring free mRNPs from productively VSV-infected cells. Figure 1 outlines the procedure used to isolate both populations of VSV mRNPs. Viral RNA was labeled in the presence of actinomycin-D by the addition of rH]uridine at 2.5 hr postinfection. At 5.5 hr, the cells were lysed and a cytoplasmic extract was prepared in hypotonic buffer containing MS+ to maintain polysome structure. A portion of this cy-
TRANSLATION
OF mRNPs
toplasmic extract was phenol extracted, and the mRNA was isolated by oligo(dT) chromatography. This mRNA was translated as a standard to which the translation of various mRNP samples could be compared. Since all samples were derived from the same cytoplasm, the amount of radioactivity in any sample indicated the amount of viral RNA recovered, and all translation reactions were adjusted to contain the same amount of VSV mRNA, either in mRNP particles or as purified mRNA. The cytoplasmic extract was fractionated by centrifugation through a sucrose gradient to separate polysome structures from free mR,NPs. Figure 2A shows an example of such a gradient. The absorbance profile at 260 nm (solid line) was used to locate the polysome structures. The majority of viral RNAs, indicated by the incorporated C’H]uridine (connected dots), was found in structures sedimenting more rapidly than 80 S monosomes. Fractions containing polysomes and those representing free mRNPs, sedimenting more slowly than 80 S ribosomes, were pooled as indicated in Fig. 2A. The pooled gradient fractions were diluted and the polysomes and the free mRNPs were collected by centrifugation. A portion of each sample was phenol extracted and the recovered RNA was analyzed on a denaturing agarose gel, shown in Fig. 3. The RNA in lanes a and b of Fig. 3 are from the faster and slower sedimenting halves of the polysome pool shown in Fig. 2A. Both halves contain intact VSV mRNAs (12,15,18, and 31 S) and also 42 S genomic RNA, indicating that the polysome region of the gradient contains other cosedimenting viral RNA structures in addition to VSV mRNA-containing polysomes. Such structures include VSV nucleocapsids and replicating and transcribing complexes. The free mRNP sample (Fig. 3, lane c) contains only four of the five VSV mRNA species and no genomic RNA. The 31 S L mRNA was not recovered in this sample because its mRNP particle sediments too closely to the 80 S monosome peak in the sucrose gradient to be easily separated from polysomes. The other four smaller VSV mRNAs (the 18 S G, 15 S N,
AFTER
POLIO
421
INFECTION
abc
de
-
Origin
- -42s
Genome
$3” ‘$S
.12s NS+M
I
1
Mg+’ Gradlent
EOTA Grodient
FIG. 3. Denaturing agarose gel analysis of VSV RNA from mRNP samples obtained during isolation. RNA was phenol extracted and ethanol precipitated from a fraction of the mRNP samples obtained from the sucrose gradients shown in Fig. 2, and analyzed on a methyl mercury hydroxide-containing agarose gel, as described under Methods. Lanes a, b, and c show RNA from samples of the gradient shown in Fig. 2A. Lanes a and b are RNA from the fast and slowly sedimenting halves (respectively) of the polysome peak shown in Fig. 2A. Lane c is RNA from the free mRNP peak of the gradient in Fig. 2A. Lane d shows the RNA from the polysome-released mRNP sample collected from the sucrose gradient shown in Fig. 2B. Lane e shows the RNA from the free mRNP sample collected from the sucrose gradient shown in Fig. 2C.
and 12 S M and NS mRNAs) sediment closely enough together to be easily collected in one pooled sample. To release mRNPs from polysomes and to separate them from VSV nucleocapsid structures, the samples were adjusted to 20 mM EDTA and fractionated again through sucrose gradients containing EDTA. The absorbance profiles demonstrate that the polysomes have been disrupted, yielding 40 and 60 S ribosomal sub-
422
JONES
AND
units (Figs. 2B, C). The radioactivity from the disrupted polysome sample was resolved into two peaks. The more slowly sedimenting peak overlaps that of the 40 S ribosomal subunit as one would expect of mRNP particles containing mRNAs of 12-18 S. The fractions containing this peak of polysome released mRNPs were pooled, and when RNA phenol extracted from this sample was run on a denaturing agarose gel, it was seen to contain the four smaller VSV mRNAs (18 S G, 15 S N, and 12 S M and NS mRNAs), as shown in Fig. 3, lane d. The faster sedimenting peak from the EDTA-containing sucrose gradient fractionation of the polysome sample was found to contain the five VSV mRNAs and also the 42 S genomic RNA when the RNA was phenol extracted and run on denaturing agarose gels (data not shown). The nature of the structures containing these RNAs in this faster sedimenting peak is not clear. When the free mRNP sample (from the M$+-containing sucrose gradient) was analyzed in the EDTA-containing sucrose gradient (Fig. 2C), the only peak of incorporated [3H]uridine seen comigrated with the 40 S ribosomal subunit, similar to the slowly sedimenting peak from the polysome sample (Fig. 2B). The fractions containing this peak were pooled and when the RNA was phenol extracted from a fraction of this pooled peak, it was found to contain the same VSV mRNA species (Fig. 3, lane e) as had the corresponding polysome-released mRNPs (compare lanes d and e, Fig. 3). The absorbance profiles in Figs. 2B and C indicate substantial amounts of 40 S ribosomal subunits contaminating the mRNP samples; however, most of the 60 S ribosomal subunits were removed, and there is no contamination by VSV nucleocapsids as indicated by the absence of 42 S genomic RNA in the RNA from the mRNP samples displayed in Fig. 3, lanes d and e. A large proportion of the protein isolated with mRNPs can be removed by treatment with 0.5 M KCl. Some investigators routinely incorporate a salt wash in their mRNP isolation procedure as a means of
EHRENFELD
removing nonspecifically associated protein contaminants. To determine whether high-salt treatment affected the translation of mRNPs, half of each pooled mRNP preparation (indicated by the horizontal bars in Figs. 2B, C) was adjusted to 0.5 M KCl, diluted in buffer containing 0.5 MKCI, and collected by centrifugation. The remaining halves were diluted with the lowsalt buffer already present in the sucrose gradient pools and were collected similarly. The resulting mRNP pellets were rinsed, resuspended in a small volume, and stored for translation. The above isolation procedure yielded four mRNP preparations: polysomal mRNPs, salt-washed polysomal mRNPs, free mRNPs, and salt-washed free RNPs. All samples were isolated in parallel from the same VSV-infected cytoplasmic extract. Analysis of the polypeptide composition of these mRNP preparations by Coomassie blue staining of SDS-PAGE gels showed a complex mixture of approximately 30 bands in all samples (data not shown). Since efforts were not made to purify the mRNPs from contaminating ribosomal subunits or other cosedimenting material, no significance could be attached to the polypeptide composition. VSV mRNP preparations contained the same spectrum of polypeptides as parallel preparations from uninfected HeLa cells, with the notable addition of one major band which migrated at the position of the VSV N protein. The presence of N protein in VSV mRNP preparations has been previously reported (Grubman and Shafritz, 1977). In all cases, the salt-washed mRNPs showed the same polypeptide composition as the corresponding non-salt-washed mRNPs, although the ratio of protein/RNA was significantly decreased by the salt wash. Translation of mRNPs. Equal amounts of VSV mRNA in purified mRNA and mRNP samples were translated in a micrococcal nuclease-treated HeLa cell SlO extract. The products of translation were separated by SDS-PAGE, and the autoradiogram is shown in Fig. 4. All four mRNP samples stimulated the synthesis of VSV polypeptides (lanes c-f) with at least the same efficiency as purified mRNA
TRANSLATION b
0
c
d
e
OF mRNPs AFTER POLIO INFECTION
centrations of the cap analog m7GMP (data not shown). All mRNP preparations were known to be contaminated with 40 S ribosomal subunits, due to the sedimentation coefficients of VSV mRNPs (Figs. 2B, C). The presence of these subunits was shown to have no effect on the translation reaction, however, by a reconstitution experiment in which 40 S subunits, prepared from uninfected HeLa cells by the same procedure as was used to isolate the VSV mRNPs, were added to translation reactions containing VSV mRNA. No effect of the HeLa cell contaminants which cosedimented with VSV mRNPs was observed.
f
-L
-G 7
-M
_
mRNA CYt
mANP Polys
mRNP PTys
mRNP Free -
mRNP F:ee
423
Message Source SW
FIG. 4. In v&o translation of VSV polysome-released and free mRNPs. Translation was performed in a HeLa cell SlO extract as described under Methods. The mRNA content of each translation reaction (except lane a) was the same as determined by the [3H]uridine-labeled RNA in the mRNP and mRNA samples (see text). Lane a, no mRNA; lane b, purified cytoplasmic mRNA; lanes c and d, polysome-released mRNPs, without (c) and with (d) salt-wash treatment; lanes e and f, free mRNPs without (e) and with (f) salt-wash treatment. The mobilities of the VSV polypeptides were determined by coelectrophoresis of purified virions, and are indicated on the right.
(lane b). Salt-washed mRNP samples (SWmRNPs) were translated with greater efficiency than the corresponding non-SW mRNPs (lanes c-d and e-f). No differences were observed in the translation of free versus polysomal mRNPs, indicating that the proteins associated with free VSV mRNPs do not act to prevent translation of those mRNPs by sequestering them into a pool distinct from that of translating mRNPs. The general characteristics of the translation of mRNPs were very similar to those of mRNA with regard to mRNP concentration dependence and potassium and magnesium ion optima (78 mMand 2.5 mM, respectively’). In addition, mRNP translation was inhibited >90% by 2 mM con-
Translation of VW mRNPs in PoliovirusInfected Cell Extracts Purified VSV mRNA is not translated in extracts prepared from poliovirus-infected cells, presumably due to a polio-induced defect in cap-binding protein functions. The addition of cap-binding protein(s) or complexes containing cap-binding protein(s) to a polio-infected cell extract can restore the ability to translate VSV mRNA (Trachsel et aL, 1980). To determine whether VSV mRNPs contained a functional cap-binding activity bound to the mRNA which could overcome the defect in extracts from polio-infected cells, we attempted to translate isolated VSV mRNPs in a polio-infected SlO extract. Since no differences had been detected between free and polysomal mRNPs, all subsequent mRNP samples were prepared by direct addition of EDTA to the VSV-infected cytoplasmic extracts followed by sedimentation in an EDTA-containing sucrose gradient. Thus, the initial fractionation into polysomal and free mRNPs (Fig. 2A) was omitted. The polio-infected HeLa cell extracts were prepared at 3l/2 hr postinfection, when host cell protein synthesis was completely inhibited. Figure 5 shows the results of translation of VSV mRNPs by the polio-infected cell extract. Neither mRNA (lane b) nor mRNPs were translated (lanes c and d); thus the mRNPs do not contain a cap-binding protein activity which can replace the deficiency in the po-
JONES AND EHRENFELD
424 0
b
c
d
a
f
e
N J -NS
-
mRNA
-
u
u
u
VSV mRNP mRNA VSVmRNP + --+
RSW Message SW
FIG. 5. Translation of VSV mRNPs in SlO extracts from polio-infected cells. Translation reactions each contained the same amount of mRNA in the form of purified mRNA or mRNP particles. Lanes a-d contained no added ribosomal salt wash (RSW). Lanes e-g contained added RSW from uninfected HeLa cells. Lane a, no added mRNA; lanes b and e, purified mRNA; lanes c and f, VSV mRNPs; lanes d and g, VSV salt-washed mRNPs.
ho-infected cell extract. This extract did support the translation of polio RNA (data not shown). If uninfected cell ribosomal salt wash was added to the extract as a source of cap-binding protein activity, the ability of the infected cell extract to translate VSV mRNA (lane e) and VSV mRNPs (lanes f and g) was restored. An alternative assay for functional capbinding activity in the ribosomal salt wash is a fractionated translation system composed of salt-washed ribosomes and a soluble cell fraction from uninfected HeLa cells, with ribosomal salt wash providing initiation factors from either uninfected or polio-infected cells. Previous work has
demonstrated that ribosomal salt wash from polio-infected cells does not support the translation of either cell (Helentjaris and Ehrenfeld, 1978) or VSV mRNA (Brown et al, 1980). When VSV mRNA or VSV mRNPs were translated in this fractionated system with uninfected cell ribosomal salt wash, they were translated well and directed the synthesis of VSV polypeptides (Fig. 6, lanes c-e). However, when the ribosomal salt wash was prepared from polio-infected cells, both the VSV mRNA and VSV mRNPs were translated poorly (Fig. 6, lanes f-h). These results confirm those obtained in the polioinfected cell extract, and again indicate
TRANSLATION 0
OF mRNPs
b
c
AFTER
d
POLIO
e
INFECTION
425
f
GNL NS-
M-
RSW Messoge SW
U -
P -
U U U P P P mRNA mRNP mRNP mRNAmRNP mRNP + +
FIG. 6. Translation of VSV mRNPs in a fractionated HeLa cell system in the presence of initiation factors from uninfected or poliovirus-infected cells. Each reaction contained the same amount of VSV mRNA, in the form of purified mRNA or mRNP particles. The translation system was reconstituted, as described under Methods, from salt-washed ribosomes, a soluble cell fraction, and ribosomal salt wash (RSW) as a source of initiation factors. Lanes a and b contained no mRNA; lanes c and f, purified mRNA; lanes d and g, VSV mRNPs; lanes e and h, salt-washed mRNPs. Lanes c-e were translated with RSW from uninfected HeLa cells; lanes f-h, RSW from polioinfected cells.
that functional cap-binding protein activity is absent from mRNP particles or insufficient to permit translation in vitro.
Translutiolz. of VW mRNPs from poliosuperin&ected! cells. Following poliovirus superinfection of VSV-infected cells, VSV protein synthesis ceases and VSV mRNPs accumulate in the cytoplasm. Purified mRNA extracted from the mRNPs has been shown to be translated efficiently in cell-free extracts (Ehrenfeld and Lund, 1977). These nontranslating VSV mRNPs were isolated and examined for evidence of an associated blocking factor or modi-
fication which would prevent or interfere with their translation in vitro. HeLa cells were infected with VSV, and one-half of the culture was superinfected with poliovirus at 1% hr post VSV infection. Both cultures were harvested 4 hr later, when it was determined that VSV protein synthesis had been completely inhibited, and mRNP particles were isolated as described above, with and without a final high-salt wash, to test the possibility that high salt might remove any putative blocking factors. RNA was extracted from the four mRNP samples (VSV mRNPs, VSV SW
JONES
426
AND
mRNPs, VSV/PV mRNPs, and VSV/PV SW mRNPs), and analyzed on denaturing agarose gels. Figure ‘7shows that the VSV mRNA in these samples was intact, as had been reported previously (Ehrenfeld and Lund, 1977) for the superinfected cells. Translation of VSV/PV mRNPs was
-0ri
EHRENFELD
performed in a fractionated uninfected HeLa cell system, and the products were compared with those from translation of control VSV mRNPs. Figure 8 shows that VSV mRNPs (lanes c and d) are translated as well as VSV mRNA (lane b); in addition, VSV/PV mRNPs are translated equally well (lanes e and f). Translation of all mRNPs is improved after the high-salt wash. Similar results were obtained in an unfractionated HeLa SlO translation system. These experiments showed that nontranslating mRNPs isolated from polio-infected cells contained no detectable blocking factor or other modification resulting from poliovirus infection. DISCUSSION
.G -N NS -&M
vsv -
VSV/Pv +
-
+
mRNP SW
FIG. 7. Denaturing agarose gel analysis of RNA from mRNP samples. RNA was phenol extracted from a fraction of the VSV mRNP samples isolated from productively infected and from poliovirus superinfected cells, and analyzed by electrophoresis in an agarose gel containing methyl mercury hydroxide. Lanes a and b show RNA from normal VSV mRNPs which had not been salt washed (a) or which had been salt washed (b). Lanes c and d show RNA from VSV mRNPs from polio-superinfected cells, which had not (c) or had (d) been salt washed.
mRNPs from numerous eukaryotic cells have been translated in vitro, and the efficiency of translation has generally been found to be similar to that of mRNA (e.g., Ernst and Arnstein, 19’75), with the exception of the inhibited translation of mRNPs from some developing tissues (Jenkins et al, 1978). Thus, mRNP-associated proteins do not appear to increase translation efficiency nonspecifically. Although a recent report by other investigators (Rosen et 4, 1982) described the isolation of VSV mRNPs which were not translationally active in a wheat germ extract, mRNPs utilized in this study were translated with an efficiency of translation similar to that of VSV mRNA, yielded similar products, and required similar ionic conditions for optimal translation. Grubman and Shafritz (1977) isolated VSV mRNPs and demonstrated that the VSV N protein, as well as a cellular 78K M, protein, was present, tightly bound to the VSV mRNA. The role of N protein in VSV mRNP preparations is not clear, although it was likely present in the preparations utilized in these studies. The analysis of mRNPs in translation studies generally utilizes relatively unpurified preparations, since the usual procedures of size fractionation in sucrose gradients rarely separates mRNPs from ribosomal subunits adequately, and never convincingly distinguishes proteins which
TRANSLATION 0
b
OF mRNPs
c
d
AFTER
POLIO
e
INFECTION
427
f
-J -NS
-M
-
mRNA -
“_SV mR+NPIVS~Pv
mFJlP
M;;ge
FIG. 8. Comparison of translation of normal VSV mRNPs and VSV mRNPs from polio-superinfected (VSV/PV) cells. Translation was performed in the fractionated HeLa system composed of saltwashed ribosomes, a soluble cell fraction, and ribosomal salt-wash from uninfected cells, as described under Methods. Lane a, no added mRNA, lane b, purified mRNA, lanes c and d, normal VSV mRNPs which had not (c) or had (d) received salt treatment; lanes e and f, VSV mRNPs isolated from polio-superinfected cells, which had not (e) or had (f) received salt treatment.
adventitiously contaminate mRNPs from those specifically associated with mRNA. Attempts at more vigorous purification (e.g., CsCl density-gradient centrifugation following formaldehyde fixation, affinity chromatography on oligo(dT) cellulose followed by elution with formamide or high salt at elevated temperatures, or crosslinking with ultraviolet light) generally render the mRNPs nonfunctional for translation, ‘but have been useful for structural analyses of protein composition. Cell fractionation studies have yielded a population of mRNPs free in the cyto-
plasm, not associated with polysomes. It is not clear whether this population eventually becomes polysome-associated or remains sequestered from the translational machinery, perhaps due to the associated proteins. The mRNAs isolated from free mRNPs are not always the same spectrum of species as is found in polysomes (McMullen et al, 1979), and structural studies of free and polysomal mRNPs have demonstrated differences in the protein composition. Free mRNPs consistently carry more protein than polysomal mRNPs (Barrieux et QL, 1976), but the 52K and 78K
428
JONES
AND
M, proteins are usually present in both mRNP populations (Cardelli and Pitot, 1977; Greenberg, 1980). We have observed no differences in the translation of free and polysome-associated VSV mRNPs, and thus there is no evidence for a sequestered population of VSV mRNAs whose translation is restricted. We consistently found that treatment of mRNPs with 0.5 M KC1 increased their translational activity significantly. The nature of the “inhibitor” which is removed by the salt treatment is not known, nor even whether it is specifically associated with the mRNPs or merely an interfering contaminant. The purpose of salt washing the mRNPs in these experiments was to remove any cap-binding activity which might exist in the mRNP, since similar high-salt treatments are routinely used to remove this activity from polysome preparations which contain endogenous mRNPs. However, no such cap-binding activity was detected in the mRNP preparations, as evidenced by their failure to be translated in a poliovirus-infected cell extract. The primary aim of this study was to determine whether any proteins involved in the translational discrimination between polio RNA and cellular (or VSV) mRNA were among the mRNA-associated proteins of the mRNP particle. We looked for either of two kinds of activity on the isolated mRNPs. First, since polio-infected cell extracts are depleted of cap-binding protein activity, we asked whether normal VSV mRNPs contained such an activity associated with the mRNA, by testing whether these mRNPs could bypass the restriction of translation seen in the polioinfected cell extracts with purified VSV mRNA. Isolated mRNPs were not translated in the infected cell extract, indicating that functional mRNA cap-binding activity was either not associated with the mRNP particle, or, if present, it was rapidly inactivated (Etchison et ok, 1982; Rose et al, 19’78)and unable to overcome the deficiency expressed in these extracts. Secondly, we examined VSV mRNPs from polio-superinfected cells for evidence of a modification or a blocking factor associated with the
EHRENFELD
mRNA which would prevent their translation in vitro. We found no evidence for such a blocking factor; there was no difference in the translation of VSV mRNPs isolated from productively infected cells or from polio-superinfected cells. Thus, unless rapid exchange of mRNP proteins occurred in vitro so as to restore an inactive mRNP into an active one, there appears to be no involvement of the mRNA-associated proteins in the inhibition of VSV translation by poliovirus. Any mRNP components which were lost during our isolation procedures would, of course, remain undetected. In an attempt to minimize the likelihood that our isolation procedures were inactivating or otherwise affecting possible mRNP-associated factors, several modifications of the isolation scheme were performed. VSV mRNPs were isolated in the absence of EDTA, and in the absence of the detergent, NP-40. In neither case were the resulting mRNP preparations translated in polio-infected cell extracts (data not shown). The functional form and subcellular location of the cap-binding protein activity, which is altered following poliovirus infection, is still unclear. Tahara et al. (1981) and Hansen et al. (1982a) have reported at least two forms of cap-binding protein complexes, isolated from the ribosomal salt wash, with different sedimentation properties, only one of which appears able to restore translation of capped mRNAs in a polio-infected cell extract. In addition, large amounts of cap-binding protein are present in the soluble portion of the cytoplasm (Hansen et al, 1982b), and its function there remains unknown. By definition, the cap-binding protein or a functional complex containing the cap-binding protein binds mRNA. In this study, we attempted to locate a form of the cap-binding protein in the intracellular mRNP particle bound to mRNA. However, we were unable to detect such an activity in isolated mRNP preparations using translation in the deficient polio-infected cell extract as an assay. It may be, therefore, that the capbinding protein-mRNA interaction is a transient one.
TRANSLATION
OF mRNPs
ACKNOWLEDGMENTS This work was supported by the National Institutes of Health Grant AI-12387. C.L.J. is the recipient of a National Institutes of Health postdoctoral fellowship AI-05986.
REFERENCES BARRIEUX, A., and ROSENFELD, M. G. (1979). Nonidentity of the 48,000 dalton protein of mRNA-protein particles and the subunit of eukaryotic initiation factor 2. J. &al. Chem 254, 8087-8090. BARRIEUX, A., INGRAHAM, H. A., NYSTUL, S., and RoSENFELD, M. G. (1976). Characterization of the association of specific proteins with messenger RNA. Biochemistry 15,3523-3528. BLOBEL, G. (1972). Protein tightly bound to globin mRNA. B&hem Biophys. Res. Commun 47,88-95. BLOBEL, G. (1973). A protein of molecular weight 78,000 bound to the polyadenylate region of eukaryotic messenger RNAs. Proc. Nat. Acad Sk USA 70.924-928. BROWN, D., HANSEN, J., and EHRENFELD, E. (1980). Specificity of initiation factor preparations from poliovirus-infected cells. J. Viral 34, 573-575. CARDELLI, J., and PITOT, H. C. (1977). Isolation and characterization of rat liver free and membrane bound polysomal messenger ribonucleoprotein particles. Biochemistry 16,5127-5134. DOYLE, M., and HOLLAND, J. J. (1972). Virus-induced interference in heterologously infected HeLa cells. J. ViroL 9, 22-28. EGLY, J. M., SCHMI’IT, M., and KEMPF, J. (1976). Characterization of a protein kinase phospboprotein system in free cytoplasmic ribonucleoprotein particles of plasma cell tumors. Biochim Biophys. A& 454, 549-557. EHRENFELD, E. (1982). Poliovirus-induced inhibition of host cell protein synthesis. Cell 28, 435-436. EHRENFELD, E., and LUND, H. (1977). Untranslated vesicular stomatitis virus messenger RNA after poliovirus infection. Virology 80, 297-308. EHRENFELD, E., and MANIS, S. (1978). Inhibition of 80s initiation complex formation by infection with poliovirus. J. Gen ViroL 43, 441-445. ERNST, V., and ARNSTEIN, H. R. V. (1975). Synthesis of (Y- and P-globin directed by messenger ribonucleoprotein from rabbit reticulocytes. B&him Biophys. Acta 378, 251-259. ETCHISON, D., MILBURN, S. C., EDERY, I., SONENBERG, N., and HERSHEY, J. W. B. (1982). Inhibition of HeLa cell protein synthesis following poliovirus infection correlates with the proteolysis of a 220,000 dalton polypeptide associated with eucaryotic initiation factor 3 and a cap binding protein complex. J. Biol Ckem 257, 14806-14810. FERNANDEZ-M~‘Ioz, R., and DARNELL, J. E. (1976).
AFTER
POLIO
INFECTION
429
Structural difference between the 5’ termini of viral and cellular mRNA in poliovirus-infected cells: Possible basis for the inhibition of host protein synthesis. J. Virol 13, 719-726. GREENBERG, J. R. (1980). Proteins crosslinked to messenger RNA by irradiating polyribosomes with ultraviolet light. Nucleic Acicls Res. 8, 5685-5701. GRUBMAN, M., and SHAFRITZ, D. A. (1977). Identification and characterization of messenger ribonucleoprotein complexes from vesicular stomatitis virus-infected HeLa cells. Virology 81. 1-16. HANSEN, J., ETCHISON, D., HERSHEY, J. W. B., and EHRENFELD, E. (1982a). Association of cap-binding protein with eucaryotic initiation factor 3 in initiation factor preparations from uninfected and poliovirus-infected HeLa cells. J. Viral 42,200-207. HANSEN, J. L., ETCHISON, D. O., HERSHEY, J. W. B., and EHRENFELD, E. (1982b). Localization of capbinding protein in subcellular fractions of HeLa cells. Mel CeU Biol. 2, 1639-1643. HELENTJARIS, T., and EHRENFELD, E. (1978). Control of protein synthesis in extracts from poliovirusinfected cells. I. mRNA discrimination by crude initiation factors. J. Viral 26, 510-521. HELLERMAN, J. G., and SHAFRITZ, D. A. (1975). Interaction of poly(A) and mRNA with eukaryotic initiator met-tRNAr binding factor: Identification of this activity on reticulocyte rihonucleic acid protein particles. Proc. Nat. Acad Sci. USA 72, 10211025. JENKINS, N. A., KAUMEYER, J. A., YOUNG, E. M., and RAFF, R. A. (1978). A test for masked message: The template activity of messenger ribonucleoprotein particles isolated from sea urchin eggs. Den Biol 63, 279-298. KAEMPFER, R., HOLLENDER, R., ABRAMS, W. R., and ISRAELI, R. (1978). Specific binding of messenger RNA and methionyl tRNAr’ by the same initiation factor for protein synthesis. Proc. Nat. Ad Sci. USA 75, 209-213. KAUFMANN, Y., GOLDSTEIN, E., and PENMAN, S. (1976). Poliovirus-induced inhibition of polypeptide initiation in vitro on native polysomes. Proc Nat. Acad Sci USA 73,X34-1838. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lmdm) 227, 680-685. LEE, K. A. W., and SONENBERG,N. (1982). Inactivation of cap-binding proteins and the shut-off of host protein synthesis by poliovirus. Proc. Nat. Acad Ski. USA 79, 3447-3451. LIAUTARD, J. P. (1977). Proteins of the polysomal messenger ribonucleoprotein are responsible for its association with the 40s ribosomal subunit in HeLa cells. Biochim. Biophys. Acta 476, 238-252. LIEBOWITZ, R., and PENMAN, S. (1971). Regulation of protein synthesis in HeLa cells. III. Inhibition during poliovirus infection. J. ViroL 8. 661-668.
430
JONES
AND
EHRENFELD
MCMULLEN, M. D., SHAW, P. H., and MARTIN, T. E. in recognition of eukaryotic mRNA by initiation factor IF-Ma. Nature (LcmdDn) 261, 291-294. (1979). Characterization of poly(A)+ RNA in free messenger ribonucleoprotein and polysomes of SONENBERG, N., MORGAN,M. A., MERRICK,W. C., and SHATKIN, A. J. (1978). A polypeptide in eukaryotic mouse taper ascites cells. J. Mol. Biol. 132,679-694. initiation factors that crosslinks specifically to the PELHAM, H. R. B., and JACKSON,R. J. (1976). An efS’terminal cap in mRNA. Proc. Nat. Ad Sci USA ficient mRNA-dependent translation system from 75, 4843-4847. reticulocyte lysates. Eur. J. Biochem. 67, 247-256. N., RUPPRECHT,K. M., HECHT, S. M., and PENMAN,S., SCHERRER,K., BECKER,Y., and DARNELL, SONENBERG, SHATKIN, A. J. (1979). Eucaryotic mRNA cap-bindJ. E. (1963).Polyribosomes in normal and poliovirusing protein: Purification by affinity chromatography infected HeLa cells and their relationship to meson Sepharose-coupled m7GDP. Proc. Nat. Acad Sci Sci USA 49,654-662. senger RNA. Proc. Nat. Ad USA 76,4345-4348. ROSE,J. K., TRACHSEL,H., LEONG,K., and BALTIMORE, SPIRIN,A. S. (1969). Informosomes. Eur. J. Biochem D. (1978) Proc. Nat. Ad Sci. USA 75, 2732-2736. 10, 20-35. ROSE, K. M., JACOB, S. T., and KUMAR, A. (1979). TAHARA, S. M., MORGAN,M. A., and SHATKIN, A. J. Poly(A) polymerase and poly(A)-specific mRNA (1981). Two forms of purified m’G-cap binding probinding protein are antigenically related. Nature teins with different effects on capped mRNA trans(L.mdm) 279, 260-262. lation in extracts of uninfected and poliovirus-inROSEN,C. A., ENNIS, H. L., and COHEN,P. S. (1982). fected cells. J. Biol Chem. 256, 7691-7694. Translational control of vesicular stomatitis virus TRACHSEL,H., SONENBERG, N., SHATKIN, A. J., ROSE, protein synthesis: Isolation of an mRNA-sequesJ. K., LEONG,K., BERGMANN,J. E., GORDON,J., and tering particle. J. Wol 44, 932-938. BALTIMORE,D. (1980). Purification of a factor that restores translation of vesicular stomatitis virus SHAF’RITZ,D. A., WEINSTEIN,J. A., SAFER,B., MERRICK, mRNA in extracts from poliovirus-infected HeLa W. C., WEBER,L. A., HICKEY, E. D., and BAGLIONI, cells. Proc. Nat. Acad Sci. USA 77, 770-774. C. (1976).Evidence for role of m7G5-phosphategroup
129.415-430
(1983)
The Effect of Poliovirus Infection on the Translation In Vitro of VSV Messenger Ribonucleoprotein Particles CHARLOTTE L. JONES
AND
ELLIE EHRENFELD’
Departments of Cellular, Viral, and Molecular Biology and Biochemistrg University of Utah School of Medicine, Salt L.&e City, Utah 85152 Received April
7, 1985;accepted June 20, 1989
Messenger ribonucleoprotein (mRNP) particles were isolated and examined for the presence of factors involved in the inhibition of protein synthesis induced by poliovirus infection. Vesicular stomatitis virus (VSV) mRNPs were used as a model for cellular mRNPs. These mRNPs were translated in HeLa cell extracts with a similar efficiency and optimal conditions to that of purified mRNA, but they were not translated in extracts prepared from poliovirus-infected HeLa cells, which have been shown to be defective in cap-binding protein activity. We conclude that mRNP proteins do not include cap-binding protein activity, since the mRNPs were not able to bypass the restriction on translation of capped mRNAs in polio-infected cell extracts. In addition, VSV mRNPs were isolated from polio-superinfected cells, in which their translation was inhibited. These mRNPs were translated in vitro as well as normal VSV mRNPs. No evidence of a modification or a blocking factor on the mRNPs which prevented their translation following polio infection was observed. Thus, within the limits of the in vitro translation assays used, no factors involved in the discrimination between polioviral and cellular or VSV mRNA could be detected in the mRNP particle. INTRODUCTION
Poliovirus infection of HeLa cells results in an inhibition of host cell protein synthesis demonstrated to occur at the initiation step of translation (Ehrenfeld and Manis, 1978; Kaufman et al, 1976; Liebowitz and Penman, 1971). Extracts from poliovirus-infected cells are unable to translate capped mRNAs in vitro (Rose et aL, 1978; Sonenberg et aL, 1979), and several investigators have reported alterations in the properties of cap-binding protein complexes isolated from infected cells (Etchison et aL, 1982; Hansen et aL, 1982a). It is likely that a virus-induced defect in the function of the cap-recognition system is responsible for the failure of cellular mRNA to bind to ribosomes and initiate translation (Blobel, 1973). ’ Present address: Department of Cellular and Developmental Biology, Biosciences West Building, University of Arizona, Tucson, Arizona 85721. ‘To whom reprint requests should be addressed. 415
The population of host cell mRNA present in infected cells has been shown to remain structurally intact and fully functional for translation in vitro (Ehrenfeld and Lund, 1977; Fernandez-Munoz and Darnell, 1976), despite its failure to be utilized in the infected cell. However, it has been shown that eukaryotic mRNA is associated in the cell with proteins to form a messenger ribonucleoprotein particle (mRNP) (Greenberg, 1980; Spirin, 1969). Numerous studies have been conducted in an attempt to understand the structural and functional role of the proteins in mRNP particles on translation of mRNA, either as inhibitory or blocking proteins which might prevent translation under specific physiologic conditions, or as structural regulators which may affect the secondary structure of mRNA so as to expose or facilitate the function of ribosomebinding sites. The purpose of this study was to determine whether any factors involved in the translational discrimination between polioviral and cellular mRNAs in 0042~6822183$3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
416
JONES
AND
the infected cell occur in the mRNP particle. A number of mRNP proteins have been described, but there is little consensus on either the number or molecular weights of these proteins, which vary with the method of isolation and the system being analyzed. Recently, ultraviolet light-induced crosslinking has been successfully utilized to identify proteins physically associated with mRNA (Greenberg, 1980). Two tightly bound proteins, 52K and 78K M,, have been identified in most systems examined (Blobel, 1972; Blobel, 1973; Greenberg, 1980). The latter protein is preferentially associated with the poly(A) sequence of the mRNA (Blobel, 1973; Greenberg, 1980), and cross-reacts with antibodies made against the poly(A) polymerase (Rose et al, 1979). Protein kinase activity has been found associated with mRNPs in a number of systems (e.g., Egly et a& 1976). Beyond that, little is known regarding the function of the mRNP proteins, but there are several situations which suggest that mRNA-associated proteins can exert both positive and negative regulatory effects. Although none of the characterized eukaryotic initiation factors have been identified among the tightly bound mRNP proteins (Barrieux and Rosenfeld, 1979), a number of initiation factors have been shown to bind mRNA in vitro. These include eIF2 (Hellerman and Shafritz, 1975), eIF4B (Shafritz et al, 1976), eIF5 (Shafritz, et al., 1976), and the cap-binding proteins (Sonenberg et al, 1978). Other activities found associated with mRNPs include a met-tRNAf binding protein (Barrieux and Rosenfeld, 1979; Hellerman and Shafritz, 1975; Kaempfer et al, 1978) and a factor which supports the binding of isolated mRNPs to the 40 S ribosomal subunit from HeLa cells (Liautard, 1979). In several differentiating or developmentally regulated systems, the proteins associated with mRNP particles have been shown to exert a negative regulation of translation. For example, mRNPs isolated from unfertilized sea urchin eggs, which are not translated in vivo, are also not translated in vitro. However, the mRNA extracted from these particles is translated
EHRENFELD
well in vitro. Both mRNA and mRNPs from fertilized eggs are translated efficiently (Jenkins et aZ., 1978). To examine the role of mRNP proteins in polio-infected cells, we have used vesicular stomatitis virus (VSV) mRNPs as a model for cellular mRNPs, since VSV protein synthesis is inhibited by poliovirus infection in the same manner as is host cell translation (Doyle and Holland, 1972; Ehrenfeld and Lund, 1977). In addition, the VSV mRNAs and polypeptides are readily identified by gel electrophoresis, and Grubman and Shafritz have previously characterized the VSV mRNP particles (Grubman and Shafritz, 1977). The studies described in this report address the question of whether functional mRNP particles contain the cap-recognition activity, bound to mRNA. Such an activity would allow mRNPs to bypass the restriction imposed on the translation of capped mRNAs in a polio-infected cell extract, which is depleted of a functional capbinding protein activity. We show that VSV mRNPs, although translationally active, are not translated in a polio-infected cell extract. In addition, VSV mRNPs isolated from a poliovirus superinfected cell contain no detectable alteration or blocking activity which prevents their translation in vitro. MATERIALS
AND
METHODS
Virus and cells. HeLa S-3 cells were grown in spinner culture in Eagle’s minimal essential medium (Joklik’s modification) supplemented with 10% heat-inactivated calf serum. Vesicular stomatitis virus (VSV) stocks were grown in HeLa cells and purified as follows: Approximately 3 X 10’ cells (5 X 106/ml) in medium without serum were infected at a multiplicity of infection of l3 PFU/cell, and the virus was adsorbed for 1 hr at 37”. At this time the cells were diluted to a concentration of 106/ml and the medium was made 2% serum. The infected cells were incubated 18 hr at 34” and the medium was then cleared of cells and cell debris by centrifugation at 800 g for 15 min. The cleared supernatant was
TRANSLATION
OF mRNPs
made 2.2% NaCl and 7% polyethylene glyco1 6000 and stirred on ice for 5-6 hr. The virus precipitate was collected by centrifugation at 1800 g for 10 min at 4” and resuspended in 6-8 ml ET buffer (10 mM Tris-Cl, 1 mM EDTA, pH ‘7.4), sonicated 2-4 min in a Melter ultrasonic water bath, and layered on six lo-50% (w/w) sucrose gradients in ET buffer. The gradients were spun for 13 hr at 26,000 rpm, 4”, in a Beckman SW 28 rotor and the visible virus bands were collected with a syringe. The virus was diluted 3-4 X with cold ET buffer and collected in pellets by centrifugation in the SW 28 rotor at 23,000 rpm for 2 hr at 4”. The virus pellets were resuspended in ET, briefly sonicated, and layered on three 5-20% sucrose (w/w in ET buffer) gradients. The gradients were spun in the SW 28 rotor at 18,000 rpm for 50 min at 4” and the virus collected, diluted, and pelleted as before. The virus pellets were resuspended in l-2 ml ET containing 10% DMSO, and aliquots were stored at -70”. Virus titers were obtained by counting plaques on monolayer HeLa cells in Eagle’s minimal essential medium supplemented with 2% heat-inactivated calf serum. Poliovirus (Mahoney strain) type 1 was grown in HeLa S-3 cells in spinner culture. Cells were infected at a concentration of 5 X lo6 cells/ml with a multiplicity of infection of 20 PFU/cell, and the virus was adsorbed for 30 min at 37”. After adsorption, serum was added to 2% and the culture was incubated at 37” for another 5% hr. The cells were then harvested by centrifugation, washed twice with Earle’s salts solution, and resuspended in RSB (10 mM Tris-Cl pH 7.4, 10 mM NaCl, 1.5 mM MgClJ. The cells were broken by freezethawing three times and NP-40 was added to 1%. Nuc1e.i were removed by 5 min centrifugation at 1300 g and the supernatant was cleared by centrifugation at 10,000 g for 15 min. This supernatant was then diluted with RSB and made 1% SDS for a final volume of 30 ml, and the virus was collected in a pellet by centrifugation in an angle 50.2 Ti rotor at 30,000 rpm for 3 hr at 20”. Pellets were resuspended by stirring on ice with a magnetic stir bar in 4.5 ml of 10 mM Tris-Cl, pH 7.4,lO mM NaCl,
AFTER
POLIO
INFECTION
417
1 mM EDTA, 1% Brij 58, and 2.25 g CsCl was added to make a final density of 1.33 g/ml. The virus was then banded by spinning in an SW 50.1 rotor at 40,000 rpm, 17 hr, 4”. The visible virus band was removed with a pipet and stored at 4”. The virus was quantitated by ultraviolet absorbance at a wavelength of 260 nm where 1 absorbance unit = 1.3 X 1013 particles/ml with a particle/PFU ratio of 100. Plaque assay and RNA synthesis rates confirmed the calculated titer. mRNP isolations. An outline of the procedure is shown in Fig. 1. HeLa cells (5 X lo* cells/100 ml) were infected with VSV at a multiplicity of lo-20 PFU/cell at 37”. After 1 hr serum was added to 2%. AT 2.5 hr, actinomycin D was added to 2 pg/ml and [3H]uridine (New England Nuclear, 42 Ci/mmol) was added to 25 &i/ml. At 5.5 hr postinfection, the cells were collected, washed three times with Earle’s salts solution, and resuspended in 3 ml RSB. NP40 was added to 0.5%; the nuclei were removed by centrifugation and washed with 1 ml RSB and 0.5% NP-40 and the wash was added to the cytoplasmic extract. The cytoplasmic extracts were loaded directly on a 7-47% (w/w) sucrose gradient in RSB, and spun in a Beckman SW 28 rotor at 16,000 rpm for 15 hr at 4” (35,000 g). The gradient was collected into 30 fractions and small samples of each fraction were analyzed for TCA-precipitable radioactivity. The appropriate fractions for polysomes and free mRNPs were pooled as indicated in Fig. 2A and diluted approximately twofold with RSB. Polysomes and mRNPs were pelleted by centrifugation in a Beckman angle 50 Ti rotor at 48,000 rpm (160,000 g) for 24 hr at 4”. The pellets were each suspended in 0.5 ml NEB (0.15 MNaCl, 10 mM Tris, 10 mM EDTA, pH 7.4) by stirring with a magnetic stir bar on ice for 2-3 hr. The samples were then each layered on 1530% sucrose (w/w) gradients in NEB and spun in the Beckman SW 28 rotor at 25,000 rpm for 14.5 hr at 4” (85,000 g). Gradients were collected as before. The mRNP-containing fractions were pooled as indicated (Figs. 2B, C). Each pooled sample was divided in half; one half was made 0.5 MKCl and diluted approximately twofold with
JONES
418
AND
EHRENFELD
CYTOPLASM
SUCROSE
VELOCITY
GRADIENT
‘FREE
POLY SOMES
t
PELLETED
PELLETED
SUCROSE
IN
POLY SOt44L
r
PELLETED
SUCROSE
VELOCITY
GRADIENT
WASHED
VELOCITY IN EDTA
FREE mRNP ‘S
I 1 t
WASHED
IN
0.5
M KC1 3 PELLETED
t
t
POLYSOMAL
POLYSOKAL
mRNP ’ S
GRADIENT
EDTA
mRNP ‘5
0.5
mRNP’S
PELLETED
FREE
mRNP’S
IN
M KC1
PELLETED
FREE mRNP’S
SW
mRNP ‘S SW
FIG. 1. Scheme for isolation
of VSV mRNPs. Details
HSB (0.5 M KCl, 3 mM MgClz, 50 m&f Tris, pH 7.4); the other half was simply diluted twofold with NEB. The mRNPs were pelleted as before, the pellets rinsed gently and resuspended as before in 0.1 ml 10 mM Tris, 10 m&f KCl, pH 7.4, and the samples stored frozen at -70”. All solutions and glassware were autoclaved and the samples were always kept cold Gel analysis ofRNA. RNA was extracted with phenol from mRNP preparations or from cytoplasmic extracts and precipitated with 2.5 vol of 95% ethanol overnight at -70”. RNA was pelleted in an Eppendorf airfuge for 5 min (150,000 9) and the pellets were dissolved in 25 ~1 HzO. To each sample was added 25 ~1 sample buffer (borate buffer, 20% glycerol, 0.005% bromophenol blue, 10 mM methyl mercury hydroxide) and incubated 15 min at room temperature. The samples were then loaded on a 1% agarose horizontal slab gel in borate buffer (0.05 M boric acid, 5 mM sodium borate, 10 mM sodium sulfate, 1 mM disodium EDTA) which was 5 mM methyl mercury hydroxide. The gel was run at 90 V with
are described
under Methods.
the borate buffer in buffer trays recirculated. The gel was soaked in 500 ml 10% acetic acid, 10 m&f cysteine for 15 min to complex the mercury. Two methanol washes of at least an hour each followed, and the methanol was pulled from the gel under vacuum for 15 min. The gel was soaked 2 hr in PPO in methanol (32 g PPO/ 250 ml methanol), rinsed three times with Hz0 for 5 min each, sealed in Saran Wrap, and exposed to film at -70”. Oligo(dT) cellulose chromutography. RNA samples were dissolved in 0.5 M KCl, 10 mMTris, pH 7.4, at a concentration of less than 0.5 mg/ml and fractionated over a 0.5-g oligo(dT) cellulose (Collaborative Research, Inc.) column equilibrated with 0.5 M KCl, 10 mlM Tris-HCl, pH 7.4. The column was then washed with 5 ml of the same buffer and the bound fraction was eluted with 10 mlM Tris-HCl, pH 7.4. RNA samples were then ethanol precipitated for further use. Translation systems. Two translation systems were used: an unfractionated HeLa SlO extract, prepared from either
5
FRACTION
IO
NUMBER
I5
20
25 FRACTION
NUMBER
TOP
5
I,,,,
l,f,,,!,!
FRACTION
IO
t
I5
I.*.-.
NUMBER
I
\
t
I,,,,, 20
25 TOP
and free mRNPs. Cytoplasmic extracts of VSV-infected to maintain polysome structure. Fractions indicated by The polysome sample from (A) was made 20 m&f EDTA as described under Methods. The horizontal bar indicates (A) was made 20 mM EDTA and sedimented as in panel centrifugation. The solid line indicates the absorbance at
BOTTOM
BOTTOM
TOP
,I,,, 0
30
3-
4-
o!
6-
65-
7-
7-
9-
io- c
e-
4
FIG. 2. Sucrose gradient fractionation for isolation of VSV mRNPs. (A) Separation of polysomal cells were sedimented through a sucrose gradient containing Mg+‘, as described under Methods, horizontal bars were pooled, and the mRNPs from each pool were collected by centrifugation. (B) to dissociate ribosomes and mRNPs, and was fractionated in a sucrose gradient containing EDTA, those fractions pooled to obtain the polysome-released mRNPs. (C) The free mRNP sample from B. The fractions indicated by the bar were pooled to obtain free mRNPs, which were collected by 260 nm; the closed circles show the [3H]uridine-labeled RNA in 0.1 ml of each 1.3-ml fraction.
BOTTOM
0
E ”
c) 0 8.
40s t
‘8
3
60s :
420
JONES
AND
uninfected or poliovirus-infected cells, and a fractionated HeLa cell system composed of salt-washed ribosomes, the ribosomal salt wash, and a soluble cell fraction. To prepare HeLa cell SlO extracts, cells were swollen in 2 cell vol of PSB (10 mM KCl, 2.5 mM DTT, 1.2 mM Mg acetate, 20 mM HEPES, pH 7.4) on ice for 10 min. Dounce homogenization (20 strokes) broke the cells, centrifugation at ‘700g for 4 min removed the nuclei, and centrifugation at 10,000g for 10 min cleared the cytoplasmic extract. This extract was then treated with micrococcal nuclease at a concentration of 20 pg/ml in 2 mM CaCla for 10 min at 18” according to the method of Pelham and Jackson (1976). The nuclease reaction was stopped by adding EGTA to 4 mM and the extract was immediately chilled and filtered through a Sephadex G-25 column (15 x 1.5 cm) equilibrated with PSB. Aliquots of the excluded material were stored frozen at -70’. Extracts from polio-infected cells were prepared in the same manner from cells which had been infected as described above and harvested 3’/2 hr postinfection. For the fractionated cell system, HeLa cell extracts were prepared and nuclease treated as described above; however, after the nuclease treatment, ribosomes were collected by centrifugation in a Beckman angle 50 Ti rotor at 48,000 rpm for 90 min at 4”. The supernatant (soluble cell fraction) was saved and concentrated (see below). The ribosomal pellets were then resuspended in PSB at a concentration of 250 AacO/ml on ice with a magnetic stir bar. To remove the initiation factors from the ribosomes, 2 M KC1 was added to 0.5 M and they were stirred 10 min on ice, after which the ribosomes were separated from the ribosomal salt wash by centrifugation in an angle 50 Ti rotor at 48,000 rpm, 4”, 90 min, and the ribosomal pellet was resuspended to give a concentration of 250 AzW/ml in 0.25 M sucrose, 0.2 mM EDTA, 1 mM DTT. Aliquots of the salt-washed ribosomes were stored at -70”. For a source of initiation factors, the ribosomal salt wash was dialyzed against 20 mMTris, pH 7.4,lOO mM KCl, 0.2 mM EDTA, 7 mM P-mercaptoethanol, and 5% glycerol, and aliquots were stored frozen at -70”.
EHRENFELD
The postribosomal supernatant was diluted in 2 vol of cold 10 mM P-mercaptoethanol, and 1 M acetic acid was added dropwise until the pH was 5.0. The resulting precipitate was collected by centrifugation at 4000 g for 5 min. The pellet was dissolved in 50 mM KCl, 3 mM Mg acetate, 30 mM Tris, pH 7.4, and 10 mM P--mercaptoethanol in l/10 the original volume. The pH was readjusted to 7.4 with KOH, and this fraction was stored frozen at -70”. Translation reactions (50 ~1) contained either 40% (v/v) SlO extract or 10% (v/v) salt-washed ribosomes, 2% (v/v) pH 5 soluble fraction, and 12% (v/v) ribosomal salt wash. In addition, the reactions contained 65 mM KCl, 28 mM HEPES, pH 7.4, 2.5 mM Mg acetate, 3 mM DTT, each amino acid except methionine at 4 PM, 2.5 pg tRNA, 1 mM ATP, 0.2 mM GTP, 25 mM creatine phosphate, 3 &i [35S]methionine (Amersham, 1220 Ci/mmol), and l-2 /*g mRNA. Reactions were incubated for 1 hr at 3O”C, diluted with SDS-PAGE sample buffer, and analyzed on 10% polyacrylamide gels (Laemmli, 1970). RESULTS
Isolation
of VSV rnRNPs
Following poliovirus infection, cellular mRNPs accumulate in the cytoplasm, not translated, and not associated with polysomes (Penman et ah, 1963). Poliovirus superinfection of VSV-infected cells produces a similar accumulation of untranslated VSV mRNPs (unpublished observations). To compare VSV mRNPs present in polioinfected cells with functional VSV mRNPs, it was first necessary to examine both the polysome-associated and the naturally occurring free mRNPs from productively VSV-infected cells. Figure 1 outlines the procedure used to isolate both populations of VSV mRNPs. Viral RNA was labeled in the presence of actinomycin-D by the addition of rH]uridine at 2.5 hr postinfection. At 5.5 hr, the cells were lysed and a cytoplasmic extract was prepared in hypotonic buffer containing MS+ to maintain polysome structure. A portion of this cy-
TRANSLATION
OF mRNPs
toplasmic extract was phenol extracted, and the mRNA was isolated by oligo(dT) chromatography. This mRNA was translated as a standard to which the translation of various mRNP samples could be compared. Since all samples were derived from the same cytoplasm, the amount of radioactivity in any sample indicated the amount of viral RNA recovered, and all translation reactions were adjusted to contain the same amount of VSV mRNA, either in mRNP particles or as purified mRNA. The cytoplasmic extract was fractionated by centrifugation through a sucrose gradient to separate polysome structures from free mR,NPs. Figure 2A shows an example of such a gradient. The absorbance profile at 260 nm (solid line) was used to locate the polysome structures. The majority of viral RNAs, indicated by the incorporated C’H]uridine (connected dots), was found in structures sedimenting more rapidly than 80 S monosomes. Fractions containing polysomes and those representing free mRNPs, sedimenting more slowly than 80 S ribosomes, were pooled as indicated in Fig. 2A. The pooled gradient fractions were diluted and the polysomes and the free mRNPs were collected by centrifugation. A portion of each sample was phenol extracted and the recovered RNA was analyzed on a denaturing agarose gel, shown in Fig. 3. The RNA in lanes a and b of Fig. 3 are from the faster and slower sedimenting halves of the polysome pool shown in Fig. 2A. Both halves contain intact VSV mRNAs (12,15,18, and 31 S) and also 42 S genomic RNA, indicating that the polysome region of the gradient contains other cosedimenting viral RNA structures in addition to VSV mRNA-containing polysomes. Such structures include VSV nucleocapsids and replicating and transcribing complexes. The free mRNP sample (Fig. 3, lane c) contains only four of the five VSV mRNA species and no genomic RNA. The 31 S L mRNA was not recovered in this sample because its mRNP particle sediments too closely to the 80 S monosome peak in the sucrose gradient to be easily separated from polysomes. The other four smaller VSV mRNAs (the 18 S G, 15 S N,
AFTER
POLIO
421
INFECTION
abc
de
-
Origin
- -42s
Genome
$3” ‘$S
.12s NS+M
I
1
Mg+’ Gradlent
EOTA Grodient
FIG. 3. Denaturing agarose gel analysis of VSV RNA from mRNP samples obtained during isolation. RNA was phenol extracted and ethanol precipitated from a fraction of the mRNP samples obtained from the sucrose gradients shown in Fig. 2, and analyzed on a methyl mercury hydroxide-containing agarose gel, as described under Methods. Lanes a, b, and c show RNA from samples of the gradient shown in Fig. 2A. Lanes a and b are RNA from the fast and slowly sedimenting halves (respectively) of the polysome peak shown in Fig. 2A. Lane c is RNA from the free mRNP peak of the gradient in Fig. 2A. Lane d shows the RNA from the polysome-released mRNP sample collected from the sucrose gradient shown in Fig. 2B. Lane e shows the RNA from the free mRNP sample collected from the sucrose gradient shown in Fig. 2C.
and 12 S M and NS mRNAs) sediment closely enough together to be easily collected in one pooled sample. To release mRNPs from polysomes and to separate them from VSV nucleocapsid structures, the samples were adjusted to 20 mM EDTA and fractionated again through sucrose gradients containing EDTA. The absorbance profiles demonstrate that the polysomes have been disrupted, yielding 40 and 60 S ribosomal sub-
422
JONES
AND
units (Figs. 2B, C). The radioactivity from the disrupted polysome sample was resolved into two peaks. The more slowly sedimenting peak overlaps that of the 40 S ribosomal subunit as one would expect of mRNP particles containing mRNAs of 12-18 S. The fractions containing this peak of polysome released mRNPs were pooled, and when RNA phenol extracted from this sample was run on a denaturing agarose gel, it was seen to contain the four smaller VSV mRNAs (18 S G, 15 S N, and 12 S M and NS mRNAs), as shown in Fig. 3, lane d. The faster sedimenting peak from the EDTA-containing sucrose gradient fractionation of the polysome sample was found to contain the five VSV mRNAs and also the 42 S genomic RNA when the RNA was phenol extracted and run on denaturing agarose gels (data not shown). The nature of the structures containing these RNAs in this faster sedimenting peak is not clear. When the free mRNP sample (from the M$+-containing sucrose gradient) was analyzed in the EDTA-containing sucrose gradient (Fig. 2C), the only peak of incorporated [3H]uridine seen comigrated with the 40 S ribosomal subunit, similar to the slowly sedimenting peak from the polysome sample (Fig. 2B). The fractions containing this peak were pooled and when the RNA was phenol extracted from a fraction of this pooled peak, it was found to contain the same VSV mRNA species (Fig. 3, lane e) as had the corresponding polysome-released mRNPs (compare lanes d and e, Fig. 3). The absorbance profiles in Figs. 2B and C indicate substantial amounts of 40 S ribosomal subunits contaminating the mRNP samples; however, most of the 60 S ribosomal subunits were removed, and there is no contamination by VSV nucleocapsids as indicated by the absence of 42 S genomic RNA in the RNA from the mRNP samples displayed in Fig. 3, lanes d and e. A large proportion of the protein isolated with mRNPs can be removed by treatment with 0.5 M KCl. Some investigators routinely incorporate a salt wash in their mRNP isolation procedure as a means of
EHRENFELD
removing nonspecifically associated protein contaminants. To determine whether high-salt treatment affected the translation of mRNPs, half of each pooled mRNP preparation (indicated by the horizontal bars in Figs. 2B, C) was adjusted to 0.5 M KCl, diluted in buffer containing 0.5 MKCI, and collected by centrifugation. The remaining halves were diluted with the lowsalt buffer already present in the sucrose gradient pools and were collected similarly. The resulting mRNP pellets were rinsed, resuspended in a small volume, and stored for translation. The above isolation procedure yielded four mRNP preparations: polysomal mRNPs, salt-washed polysomal mRNPs, free mRNPs, and salt-washed free RNPs. All samples were isolated in parallel from the same VSV-infected cytoplasmic extract. Analysis of the polypeptide composition of these mRNP preparations by Coomassie blue staining of SDS-PAGE gels showed a complex mixture of approximately 30 bands in all samples (data not shown). Since efforts were not made to purify the mRNPs from contaminating ribosomal subunits or other cosedimenting material, no significance could be attached to the polypeptide composition. VSV mRNP preparations contained the same spectrum of polypeptides as parallel preparations from uninfected HeLa cells, with the notable addition of one major band which migrated at the position of the VSV N protein. The presence of N protein in VSV mRNP preparations has been previously reported (Grubman and Shafritz, 1977). In all cases, the salt-washed mRNPs showed the same polypeptide composition as the corresponding non-salt-washed mRNPs, although the ratio of protein/RNA was significantly decreased by the salt wash. Translation of mRNPs. Equal amounts of VSV mRNA in purified mRNA and mRNP samples were translated in a micrococcal nuclease-treated HeLa cell SlO extract. The products of translation were separated by SDS-PAGE, and the autoradiogram is shown in Fig. 4. All four mRNP samples stimulated the synthesis of VSV polypeptides (lanes c-f) with at least the same efficiency as purified mRNA
TRANSLATION b
0
c
d
e
OF mRNPs AFTER POLIO INFECTION
centrations of the cap analog m7GMP (data not shown). All mRNP preparations were known to be contaminated with 40 S ribosomal subunits, due to the sedimentation coefficients of VSV mRNPs (Figs. 2B, C). The presence of these subunits was shown to have no effect on the translation reaction, however, by a reconstitution experiment in which 40 S subunits, prepared from uninfected HeLa cells by the same procedure as was used to isolate the VSV mRNPs, were added to translation reactions containing VSV mRNA. No effect of the HeLa cell contaminants which cosedimented with VSV mRNPs was observed.
f
-L
-G 7
-M
_
mRNA CYt
mANP Polys
mRNP PTys
mRNP Free -
mRNP F:ee
423
Message Source SW
FIG. 4. In v&o translation of VSV polysome-released and free mRNPs. Translation was performed in a HeLa cell SlO extract as described under Methods. The mRNA content of each translation reaction (except lane a) was the same as determined by the [3H]uridine-labeled RNA in the mRNP and mRNA samples (see text). Lane a, no mRNA; lane b, purified cytoplasmic mRNA; lanes c and d, polysome-released mRNPs, without (c) and with (d) salt-wash treatment; lanes e and f, free mRNPs without (e) and with (f) salt-wash treatment. The mobilities of the VSV polypeptides were determined by coelectrophoresis of purified virions, and are indicated on the right.
(lane b). Salt-washed mRNP samples (SWmRNPs) were translated with greater efficiency than the corresponding non-SW mRNPs (lanes c-d and e-f). No differences were observed in the translation of free versus polysomal mRNPs, indicating that the proteins associated with free VSV mRNPs do not act to prevent translation of those mRNPs by sequestering them into a pool distinct from that of translating mRNPs. The general characteristics of the translation of mRNPs were very similar to those of mRNA with regard to mRNP concentration dependence and potassium and magnesium ion optima (78 mMand 2.5 mM, respectively’). In addition, mRNP translation was inhibited >90% by 2 mM con-
Translation of VW mRNPs in PoliovirusInfected Cell Extracts Purified VSV mRNA is not translated in extracts prepared from poliovirus-infected cells, presumably due to a polio-induced defect in cap-binding protein functions. The addition of cap-binding protein(s) or complexes containing cap-binding protein(s) to a polio-infected cell extract can restore the ability to translate VSV mRNA (Trachsel et aL, 1980). To determine whether VSV mRNPs contained a functional cap-binding activity bound to the mRNA which could overcome the defect in extracts from polio-infected cells, we attempted to translate isolated VSV mRNPs in a polio-infected SlO extract. Since no differences had been detected between free and polysomal mRNPs, all subsequent mRNP samples were prepared by direct addition of EDTA to the VSV-infected cytoplasmic extracts followed by sedimentation in an EDTA-containing sucrose gradient. Thus, the initial fractionation into polysomal and free mRNPs (Fig. 2A) was omitted. The polio-infected HeLa cell extracts were prepared at 3l/2 hr postinfection, when host cell protein synthesis was completely inhibited. Figure 5 shows the results of translation of VSV mRNPs by the polio-infected cell extract. Neither mRNA (lane b) nor mRNPs were translated (lanes c and d); thus the mRNPs do not contain a cap-binding protein activity which can replace the deficiency in the po-
JONES AND EHRENFELD
424 0
b
c
d
a
f
e
N J -NS
-
mRNA
-
u
u
u
VSV mRNP mRNA VSVmRNP + --+
RSW Message SW
FIG. 5. Translation of VSV mRNPs in SlO extracts from polio-infected cells. Translation reactions each contained the same amount of mRNA in the form of purified mRNA or mRNP particles. Lanes a-d contained no added ribosomal salt wash (RSW). Lanes e-g contained added RSW from uninfected HeLa cells. Lane a, no added mRNA; lanes b and e, purified mRNA; lanes c and f, VSV mRNPs; lanes d and g, VSV salt-washed mRNPs.
ho-infected cell extract. This extract did support the translation of polio RNA (data not shown). If uninfected cell ribosomal salt wash was added to the extract as a source of cap-binding protein activity, the ability of the infected cell extract to translate VSV mRNA (lane e) and VSV mRNPs (lanes f and g) was restored. An alternative assay for functional capbinding activity in the ribosomal salt wash is a fractionated translation system composed of salt-washed ribosomes and a soluble cell fraction from uninfected HeLa cells, with ribosomal salt wash providing initiation factors from either uninfected or polio-infected cells. Previous work has
demonstrated that ribosomal salt wash from polio-infected cells does not support the translation of either cell (Helentjaris and Ehrenfeld, 1978) or VSV mRNA (Brown et al, 1980). When VSV mRNA or VSV mRNPs were translated in this fractionated system with uninfected cell ribosomal salt wash, they were translated well and directed the synthesis of VSV polypeptides (Fig. 6, lanes c-e). However, when the ribosomal salt wash was prepared from polio-infected cells, both the VSV mRNA and VSV mRNPs were translated poorly (Fig. 6, lanes f-h). These results confirm those obtained in the polioinfected cell extract, and again indicate
TRANSLATION 0
OF mRNPs
b
c
AFTER
d
POLIO
e
INFECTION
425
f
GNL NS-
M-
RSW Messoge SW
U -
P -
U U U P P P mRNA mRNP mRNP mRNAmRNP mRNP + +
FIG. 6. Translation of VSV mRNPs in a fractionated HeLa cell system in the presence of initiation factors from uninfected or poliovirus-infected cells. Each reaction contained the same amount of VSV mRNA, in the form of purified mRNA or mRNP particles. The translation system was reconstituted, as described under Methods, from salt-washed ribosomes, a soluble cell fraction, and ribosomal salt wash (RSW) as a source of initiation factors. Lanes a and b contained no mRNA; lanes c and f, purified mRNA; lanes d and g, VSV mRNPs; lanes e and h, salt-washed mRNPs. Lanes c-e were translated with RSW from uninfected HeLa cells; lanes f-h, RSW from polioinfected cells.
that functional cap-binding protein activity is absent from mRNP particles or insufficient to permit translation in vitro.
Translutiolz. of VW mRNPs from poliosuperin&ected! cells. Following poliovirus superinfection of VSV-infected cells, VSV protein synthesis ceases and VSV mRNPs accumulate in the cytoplasm. Purified mRNA extracted from the mRNPs has been shown to be translated efficiently in cell-free extracts (Ehrenfeld and Lund, 1977). These nontranslating VSV mRNPs were isolated and examined for evidence of an associated blocking factor or modi-
fication which would prevent or interfere with their translation in vitro. HeLa cells were infected with VSV, and one-half of the culture was superinfected with poliovirus at 1% hr post VSV infection. Both cultures were harvested 4 hr later, when it was determined that VSV protein synthesis had been completely inhibited, and mRNP particles were isolated as described above, with and without a final high-salt wash, to test the possibility that high salt might remove any putative blocking factors. RNA was extracted from the four mRNP samples (VSV mRNPs, VSV SW
JONES
426
AND
mRNPs, VSV/PV mRNPs, and VSV/PV SW mRNPs), and analyzed on denaturing agarose gels. Figure ‘7shows that the VSV mRNA in these samples was intact, as had been reported previously (Ehrenfeld and Lund, 1977) for the superinfected cells. Translation of VSV/PV mRNPs was
-0ri
EHRENFELD
performed in a fractionated uninfected HeLa cell system, and the products were compared with those from translation of control VSV mRNPs. Figure 8 shows that VSV mRNPs (lanes c and d) are translated as well as VSV mRNA (lane b); in addition, VSV/PV mRNPs are translated equally well (lanes e and f). Translation of all mRNPs is improved after the high-salt wash. Similar results were obtained in an unfractionated HeLa SlO translation system. These experiments showed that nontranslating mRNPs isolated from polio-infected cells contained no detectable blocking factor or other modification resulting from poliovirus infection. DISCUSSION
.G -N NS -&M
vsv -
VSV/Pv +
-
+
mRNP SW
FIG. 7. Denaturing agarose gel analysis of RNA from mRNP samples. RNA was phenol extracted from a fraction of the VSV mRNP samples isolated from productively infected and from poliovirus superinfected cells, and analyzed by electrophoresis in an agarose gel containing methyl mercury hydroxide. Lanes a and b show RNA from normal VSV mRNPs which had not been salt washed (a) or which had been salt washed (b). Lanes c and d show RNA from VSV mRNPs from polio-superinfected cells, which had not (c) or had (d) been salt washed.
mRNPs from numerous eukaryotic cells have been translated in vitro, and the efficiency of translation has generally been found to be similar to that of mRNA (e.g., Ernst and Arnstein, 19’75), with the exception of the inhibited translation of mRNPs from some developing tissues (Jenkins et al, 1978). Thus, mRNP-associated proteins do not appear to increase translation efficiency nonspecifically. Although a recent report by other investigators (Rosen et 4, 1982) described the isolation of VSV mRNPs which were not translationally active in a wheat germ extract, mRNPs utilized in this study were translated with an efficiency of translation similar to that of VSV mRNA, yielded similar products, and required similar ionic conditions for optimal translation. Grubman and Shafritz (1977) isolated VSV mRNPs and demonstrated that the VSV N protein, as well as a cellular 78K M, protein, was present, tightly bound to the VSV mRNA. The role of N protein in VSV mRNP preparations is not clear, although it was likely present in the preparations utilized in these studies. The analysis of mRNPs in translation studies generally utilizes relatively unpurified preparations, since the usual procedures of size fractionation in sucrose gradients rarely separates mRNPs from ribosomal subunits adequately, and never convincingly distinguishes proteins which
TRANSLATION 0
b
OF mRNPs
c
d
AFTER
POLIO
e
INFECTION
427
f
-J -NS
-M
-
mRNA -
“_SV mR+NPIVS~Pv
mFJlP
M;;ge
FIG. 8. Comparison of translation of normal VSV mRNPs and VSV mRNPs from polio-superinfected (VSV/PV) cells. Translation was performed in the fractionated HeLa system composed of saltwashed ribosomes, a soluble cell fraction, and ribosomal salt-wash from uninfected cells, as described under Methods. Lane a, no added mRNA, lane b, purified mRNA, lanes c and d, normal VSV mRNPs which had not (c) or had (d) received salt treatment; lanes e and f, VSV mRNPs isolated from polio-superinfected cells, which had not (e) or had (f) received salt treatment.
adventitiously contaminate mRNPs from those specifically associated with mRNA. Attempts at more vigorous purification (e.g., CsCl density-gradient centrifugation following formaldehyde fixation, affinity chromatography on oligo(dT) cellulose followed by elution with formamide or high salt at elevated temperatures, or crosslinking with ultraviolet light) generally render the mRNPs nonfunctional for translation, ‘but have been useful for structural analyses of protein composition. Cell fractionation studies have yielded a population of mRNPs free in the cyto-
plasm, not associated with polysomes. It is not clear whether this population eventually becomes polysome-associated or remains sequestered from the translational machinery, perhaps due to the associated proteins. The mRNAs isolated from free mRNPs are not always the same spectrum of species as is found in polysomes (McMullen et al, 1979), and structural studies of free and polysomal mRNPs have demonstrated differences in the protein composition. Free mRNPs consistently carry more protein than polysomal mRNPs (Barrieux et QL, 1976), but the 52K and 78K
428
JONES
AND
M, proteins are usually present in both mRNP populations (Cardelli and Pitot, 1977; Greenberg, 1980). We have observed no differences in the translation of free and polysome-associated VSV mRNPs, and thus there is no evidence for a sequestered population of VSV mRNAs whose translation is restricted. We consistently found that treatment of mRNPs with 0.5 M KC1 increased their translational activity significantly. The nature of the “inhibitor” which is removed by the salt treatment is not known, nor even whether it is specifically associated with the mRNPs or merely an interfering contaminant. The purpose of salt washing the mRNPs in these experiments was to remove any cap-binding activity which might exist in the mRNP, since similar high-salt treatments are routinely used to remove this activity from polysome preparations which contain endogenous mRNPs. However, no such cap-binding activity was detected in the mRNP preparations, as evidenced by their failure to be translated in a poliovirus-infected cell extract. The primary aim of this study was to determine whether any proteins involved in the translational discrimination between polio RNA and cellular (or VSV) mRNA were among the mRNA-associated proteins of the mRNP particle. We looked for either of two kinds of activity on the isolated mRNPs. First, since polio-infected cell extracts are depleted of cap-binding protein activity, we asked whether normal VSV mRNPs contained such an activity associated with the mRNA, by testing whether these mRNPs could bypass the restriction of translation seen in the polioinfected cell extracts with purified VSV mRNA. Isolated mRNPs were not translated in the infected cell extract, indicating that functional mRNA cap-binding activity was either not associated with the mRNP particle, or, if present, it was rapidly inactivated (Etchison et ok, 1982; Rose et al, 19’78)and unable to overcome the deficiency expressed in these extracts. Secondly, we examined VSV mRNPs from polio-superinfected cells for evidence of a modification or a blocking factor associated with the
EHRENFELD
mRNA which would prevent their translation in vitro. We found no evidence for such a blocking factor; there was no difference in the translation of VSV mRNPs isolated from productively infected cells or from polio-superinfected cells. Thus, unless rapid exchange of mRNP proteins occurred in vitro so as to restore an inactive mRNP into an active one, there appears to be no involvement of the mRNA-associated proteins in the inhibition of VSV translation by poliovirus. Any mRNP components which were lost during our isolation procedures would, of course, remain undetected. In an attempt to minimize the likelihood that our isolation procedures were inactivating or otherwise affecting possible mRNP-associated factors, several modifications of the isolation scheme were performed. VSV mRNPs were isolated in the absence of EDTA, and in the absence of the detergent, NP-40. In neither case were the resulting mRNP preparations translated in polio-infected cell extracts (data not shown). The functional form and subcellular location of the cap-binding protein activity, which is altered following poliovirus infection, is still unclear. Tahara et al. (1981) and Hansen et al. (1982a) have reported at least two forms of cap-binding protein complexes, isolated from the ribosomal salt wash, with different sedimentation properties, only one of which appears able to restore translation of capped mRNAs in a polio-infected cell extract. In addition, large amounts of cap-binding protein are present in the soluble portion of the cytoplasm (Hansen et al, 1982b), and its function there remains unknown. By definition, the cap-binding protein or a functional complex containing the cap-binding protein binds mRNA. In this study, we attempted to locate a form of the cap-binding protein in the intracellular mRNP particle bound to mRNA. However, we were unable to detect such an activity in isolated mRNP preparations using translation in the deficient polio-infected cell extract as an assay. It may be, therefore, that the capbinding protein-mRNA interaction is a transient one.
TRANSLATION
OF mRNPs
ACKNOWLEDGMENTS This work was supported by the National Institutes of Health Grant AI-12387. C.L.J. is the recipient of a National Institutes of Health postdoctoral fellowship AI-05986.
REFERENCES BARRIEUX, A., and ROSENFELD, M. G. (1979). Nonidentity of the 48,000 dalton protein of mRNA-protein particles and the subunit of eukaryotic initiation factor 2. J. &al. Chem 254, 8087-8090. BARRIEUX, A., INGRAHAM, H. A., NYSTUL, S., and RoSENFELD, M. G. (1976). Characterization of the association of specific proteins with messenger RNA. Biochemistry 15,3523-3528. BLOBEL, G. (1972). Protein tightly bound to globin mRNA. B&hem Biophys. Res. Commun 47,88-95. BLOBEL, G. (1973). A protein of molecular weight 78,000 bound to the polyadenylate region of eukaryotic messenger RNAs. Proc. Nat. Acad Sk USA 70.924-928. BROWN, D., HANSEN, J., and EHRENFELD, E. (1980). Specificity of initiation factor preparations from poliovirus-infected cells. J. Viral 34, 573-575. CARDELLI, J., and PITOT, H. C. (1977). Isolation and characterization of rat liver free and membrane bound polysomal messenger ribonucleoprotein particles. Biochemistry 16,5127-5134. DOYLE, M., and HOLLAND, J. J. (1972). Virus-induced interference in heterologously infected HeLa cells. J. ViroL 9, 22-28. EGLY, J. M., SCHMI’IT, M., and KEMPF, J. (1976). Characterization of a protein kinase phospboprotein system in free cytoplasmic ribonucleoprotein particles of plasma cell tumors. Biochim Biophys. A& 454, 549-557. EHRENFELD, E. (1982). Poliovirus-induced inhibition of host cell protein synthesis. Cell 28, 435-436. EHRENFELD, E., and LUND, H. (1977). Untranslated vesicular stomatitis virus messenger RNA after poliovirus infection. Virology 80, 297-308. EHRENFELD, E., and MANIS, S. (1978). Inhibition of 80s initiation complex formation by infection with poliovirus. J. Gen ViroL 43, 441-445. ERNST, V., and ARNSTEIN, H. R. V. (1975). Synthesis of (Y- and P-globin directed by messenger ribonucleoprotein from rabbit reticulocytes. B&him Biophys. Acta 378, 251-259. ETCHISON, D., MILBURN, S. C., EDERY, I., SONENBERG, N., and HERSHEY, J. W. B. (1982). Inhibition of HeLa cell protein synthesis following poliovirus infection correlates with the proteolysis of a 220,000 dalton polypeptide associated with eucaryotic initiation factor 3 and a cap binding protein complex. J. Biol Ckem 257, 14806-14810. FERNANDEZ-M~‘Ioz, R., and DARNELL, J. E. (1976).
AFTER
POLIO
INFECTION
429
Structural difference between the 5’ termini of viral and cellular mRNA in poliovirus-infected cells: Possible basis for the inhibition of host protein synthesis. J. Virol 13, 719-726. GREENBERG, J. R. (1980). Proteins crosslinked to messenger RNA by irradiating polyribosomes with ultraviolet light. Nucleic Acicls Res. 8, 5685-5701. GRUBMAN, M., and SHAFRITZ, D. A. (1977). Identification and characterization of messenger ribonucleoprotein complexes from vesicular stomatitis virus-infected HeLa cells. Virology 81. 1-16. HANSEN, J., ETCHISON, D., HERSHEY, J. W. B., and EHRENFELD, E. (1982a). Association of cap-binding protein with eucaryotic initiation factor 3 in initiation factor preparations from uninfected and poliovirus-infected HeLa cells. J. Viral 42,200-207. HANSEN, J. L., ETCHISON, D. O., HERSHEY, J. W. B., and EHRENFELD, E. (1982b). Localization of capbinding protein in subcellular fractions of HeLa cells. Mel CeU Biol. 2, 1639-1643. HELENTJARIS, T., and EHRENFELD, E. (1978). Control of protein synthesis in extracts from poliovirusinfected cells. I. mRNA discrimination by crude initiation factors. J. Viral 26, 510-521. HELLERMAN, J. G., and SHAFRITZ, D. A. (1975). Interaction of poly(A) and mRNA with eukaryotic initiator met-tRNAr binding factor: Identification of this activity on reticulocyte rihonucleic acid protein particles. Proc. Nat. Acad Sci. USA 72, 10211025. JENKINS, N. A., KAUMEYER, J. A., YOUNG, E. M., and RAFF, R. A. (1978). A test for masked message: The template activity of messenger ribonucleoprotein particles isolated from sea urchin eggs. Den Biol 63, 279-298. KAEMPFER, R., HOLLENDER, R., ABRAMS, W. R., and ISRAELI, R. (1978). Specific binding of messenger RNA and methionyl tRNAr’ by the same initiation factor for protein synthesis. Proc. Nat. Ad Sci. USA 75, 209-213. KAUFMANN, Y., GOLDSTEIN, E., and PENMAN, S. (1976). Poliovirus-induced inhibition of polypeptide initiation in vitro on native polysomes. Proc Nat. Acad Sci USA 73,X34-1838. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lmdm) 227, 680-685. LEE, K. A. W., and SONENBERG,N. (1982). Inactivation of cap-binding proteins and the shut-off of host protein synthesis by poliovirus. Proc. Nat. Acad Ski. USA 79, 3447-3451. LIAUTARD, J. P. (1977). Proteins of the polysomal messenger ribonucleoprotein are responsible for its association with the 40s ribosomal subunit in HeLa cells. Biochim. Biophys. Acta 476, 238-252. LIEBOWITZ, R., and PENMAN, S. (1971). Regulation of protein synthesis in HeLa cells. III. Inhibition during poliovirus infection. J. ViroL 8. 661-668.
430
JONES
AND
EHRENFELD
MCMULLEN, M. D., SHAW, P. H., and MARTIN, T. E. in recognition of eukaryotic mRNA by initiation factor IF-Ma. Nature (LcmdDn) 261, 291-294. (1979). Characterization of poly(A)+ RNA in free messenger ribonucleoprotein and polysomes of SONENBERG, N., MORGAN,M. A., MERRICK,W. C., and SHATKIN, A. J. (1978). A polypeptide in eukaryotic mouse taper ascites cells. J. Mol. Biol. 132,679-694. initiation factors that crosslinks specifically to the PELHAM, H. R. B., and JACKSON,R. J. (1976). An efS’terminal cap in mRNA. Proc. Nat. Ad Sci USA ficient mRNA-dependent translation system from 75, 4843-4847. reticulocyte lysates. Eur. J. Biochem. 67, 247-256. N., RUPPRECHT,K. M., HECHT, S. M., and PENMAN,S., SCHERRER,K., BECKER,Y., and DARNELL, SONENBERG, SHATKIN, A. J. (1979). Eucaryotic mRNA cap-bindJ. E. (1963).Polyribosomes in normal and poliovirusing protein: Purification by affinity chromatography infected HeLa cells and their relationship to meson Sepharose-coupled m7GDP. Proc. Nat. Acad Sci Sci USA 49,654-662. senger RNA. Proc. Nat. Ad USA 76,4345-4348. ROSE,J. K., TRACHSEL,H., LEONG,K., and BALTIMORE, SPIRIN,A. S. (1969). Informosomes. Eur. J. Biochem D. (1978) Proc. Nat. Ad Sci. USA 75, 2732-2736. 10, 20-35. ROSE, K. M., JACOB, S. T., and KUMAR, A. (1979). TAHARA, S. M., MORGAN,M. A., and SHATKIN, A. J. Poly(A) polymerase and poly(A)-specific mRNA (1981). Two forms of purified m’G-cap binding probinding protein are antigenically related. Nature teins with different effects on capped mRNA trans(L.mdm) 279, 260-262. lation in extracts of uninfected and poliovirus-inROSEN,C. A., ENNIS, H. L., and COHEN,P. S. (1982). fected cells. J. Biol Chem. 256, 7691-7694. Translational control of vesicular stomatitis virus TRACHSEL,H., SONENBERG, N., SHATKIN, A. J., ROSE, protein synthesis: Isolation of an mRNA-sequesJ. K., LEONG,K., BERGMANN,J. E., GORDON,J., and tering particle. J. Wol 44, 932-938. BALTIMORE,D. (1980). Purification of a factor that restores translation of vesicular stomatitis virus SHAF’RITZ,D. A., WEINSTEIN,J. A., SAFER,B., MERRICK, mRNA in extracts from poliovirus-infected HeLa W. C., WEBER,L. A., HICKEY, E. D., and BAGLIONI, cells. Proc. Nat. Acad Sci. USA 77, 770-774. C. (1976).Evidence for role of m7G5-phosphategroup