- Email: [email protected]
Aminoacylation and messenger functions of eggplant mosaic virus RNA
VIROLOGY
97, 354-365 (1979)
Aminoacylation
and Messenger
Functions
of Eggplant
Mosaic
Virus RNA
T. C. HALL,*,’ MONIQUE PINCK,? Y. MA, H. M. DURANTON,S AND?‘. L. GERMAN * Department Molkulaire
of Horticulture, University of Wisconsin, Madison, Wisconsin 53706; tlnstitut o!~ Biologie at Cellulaire du CNRS, 15 rue Descartes, 67000-Strasbourg, France; and Slnstitut de Botanique, 28 rue Goethe, 6708%Straebourg, France Accepted May 1, 1979
The predominant RNA of eggplant mosaic virus (EMV) was found to have a molecular weight of 1.9 x lo6 by formamide gel electrophoresis. Electrophoretic analysis of virion protein under dissociating conditions revealed two polypeptides, a major component of 21,000 daltons, and a minor component of 22,000 daltons. Both peptides were present in translation products coded by RNA isolated from virions, but the proportion of the 22,000dalton peptide was higher in products synthesized using RNA isolated from EMV-infected Datura leaves as messenger. Since the viral RNA showed a marked tendency to aggregate, it is possible that these polypeptides were translated from trace amounts of small EMV RNAs present as contaminants of the 1.9 x 105-dalton genomic RNA. Although the addition of tRNA from infected or healthy Datura leaves, or from wheat germ, stimulated amino acid incorporation, no changes were discerned in the profile of cell-free translation products after electrophoretic separation. Nonaminoacylated and valylated EMV RNA stimulated similar levels of amino acid incorporation, and the translation products appeared identical. Valine bound to genomic EMV RNA was not donated during protein synthesis.
the 5’-end and in that the 3’-terminus acUntil recently, the tymoviruses, includ- cepts valine (Klein, 1976; Pinck et al., 1972). ing eggplant mosaic virus (EMV) and turnip Additionally, EMV RNA typically has subyellow mosaic virus (TYMV), were thought stantial quantities of associated or comto have only a single RNA component. Klein plexed tRNA (Pinck et al., 1974; Bouley et al. (1976) have described the structure et al., 1976; Pinck and Hall, 1978). In this and messenger function of two distinct study we examined the influence of added RNA components of TYMV of 2 x lo6 and tRNA and aminoacylation on translation of 0.3 x lo6 daltons, and Pleij et al. (1976) EMV RNA. We also investigated if any donation of valine from the 3’-end of the found that mild heat treatment of TYMV genome RNA occurred during protein synRNA yielded a series of RNA classes, the thesis. smallest of which was an efficient messenger for coat protein. Coat protein was also MATERIALS AND METHODS observed among the translation products Virus and RNA preparation. EMV and using unfractionated TYMV RNA as mesEMV RNA were prepared as described by senger (Benicourt and Haenni, 1976). Pinck and Hall (1978), LKB Ultrogel AcA Bouley et al. (1976) and Szybiak et al. (1978) showed that three classes of virions 22 being typically used for purification of could be separated by equilibrium cen- RNA. tRNA was extracted according to the trifugation of EMV in CsCl. Heavy (bot- procedure of Chan et al. (1976) from healthy tom component) virions predominate and and EMV-infected Datura leaves and from contain infective RNA of about 2 x lo6 wheat germ. Much of the contaminating daltons. EMV RNA is similar to TYMV DNA was removed by spooling onto a glass RNA in having an m7GpppX structure at rod; the remainder was eliminated during gel filtration on a 130 x 0.7-cm column of ’ To whom requests for reprints should be sent. AcA 54 gel. The RNA insoluble in 2 M NalNTRODUCTION
0042-6822/79/120354-12$02.00/O Copyright All rights
0 19’79 by Academic Press, Inc. of reproduction in any form reserved.
354
EMV RNA AMINOACYLATION
acetate (obtained during tRNA preparation from total nucleic acid extracts of healthy leaves) was dissolved and infected htura in water, precipitated with ethanol, redissolved in water, and filtered through a 72 x 0.7~cm column of AcA 34 gel using 10 mM Na-cacodylate buffer (pH 5.7) containing 0.1 M NaCl and 3 mM NaN,. Nucleic acid eluting as a peak of excluded material was precipitated with ethanol and subsequently used as a crude messenger (mRNA) fraction. RNA eluting at other positions showed no messenger activity in vitro. For electrophoresis, EMV RNA was dissolved in 4 M urea and an equal volume of formamide added. After heating to 60” for 5 min, the samples were quench-cooled in ice water and then applied to 3.5% polyacrylamide gels (0.6 x 10 cm) containing 99% formamide (Duesberg and Vogt, 1973). Electrophoresis at 80 V was carried out at room temperature for 17 hr. Gels were stained in 0.1% (w/v) toluidine blue in 1% acetic acid for 1 hr. Synthetase and aminoacylation reactions. These were as described by Pinck
and Hall (1978), except that the reaction volumes were scaled up to as much as 1 ml for preparative aminoacylation of tRNAs and EMV RNA fractions. Acetylation of aminoacylated RNA was according to Lapidot et al. (1967). Aminoacylated and acetylated RNA derivatives were purified by filtration (acidic buffer containing 0.1 M NaCl and 3 mM NaN, in 10 mM-Na cacodylate, pH 5.7, being used for all steps) on AcA 54 gel (for tRNAs) or AcA 34 gel (for viral RNAs), followed by precipitation with ethanol. Amino acid donation experiments were similar to those of Chen and Hall (1973). Cell-free protein synthesis and product electrophoresis. Wheat germ S23 prepara-
tion and amino acid incorporation conditions were as described by Kohl and Hall (1977), except that a French press was used to disrupt the wheat germ and [35S]Met, [3H]Leu, and [3H]Val were used as radioactive substrates. Incorporation was measured by taking 5-~1 samples of the 50-~1 reactions for assay according to the filter paper disk technique of McLeester and Hall
AND TRANSLATION
355
(1977), and reactions were stopped by freezing and storage at -20”. For product analysis, portions of the reactions were dissociated by heating (85” for 6 mm) with an equal volume of a solution (pH 6.8 with HCl) containing 10 mil! Tris, 10 mM EDTA, 100 mil4 NaCl, 4% (w/v) mercaptoethanol, and 20% (v/v) glycerol. A trace of bromophenol blue was added as tracking dye. Sample volumes (2 to 12 ~1) were chosen to contain about 50,000 cpm (by the disk assay) and applied to Laemmli (1970) gels (containing a 7-14% acrylamide gradient) and subjected to electrophoresis (anode at the bottom) in a Studier (1973) apparatus. For fluorography (Bonner and Laskey, 1974), these 0.75-mm-thick gels were soaked in two changes of DMSO (15 min each), then covered with DMSO containing 20% (v/v) PPO for 30 min, and shaken gently. On placing in water, the gel expanded to its original size within 2 min and was matched with a slice taken from the same gel containing EMV coat protein (a kind gift of Dr. D. Chollot, Strasbourg) and other standard proteins (Boehringer-Mannheim “Combithek” molecular weight standards: trypsin inhibitor, 21,500; bovine serum albumin, 68,000; Escherichia coli RNA polymerase a, 39,000; /3, 155,000; p’, 165,000) and stained with Coomassie brilliant blue (Weber and Osborn, 1969). The gel was dried onto filter paper (Studier, 1973), then exposed to Kodak No-Screen or Royal X-omat X-ray film. In later experiments, the PPO-impregnated gel was swollen in water, then stained for 3 min with a fresh solution of Coomassie brilliant blue and destained prior to drying. This procedure only slightly detracted from the efficiency of fluorography and permitted direct comparison of the migration of stained markers (mixed with reaction products) and the radioactive products. Cell-free reaction products coded by BMV RNA were also used as radioactive markers. Peptide mapping. A mixture (300 ~1) of authentic EMV coat protein (Collot et al., 1977) and [3H]Leu-labeled EMV RNAcoded translation products (1.7 x lo6 dpm) were dissociated as described above and applied to a 16% Laemmli (1970) gel. After electrophoresis, the gel was briefly stained (Weber and Osborn, 1969) to locate the
HALL ET AL.
356
22,000 and 21,000 MW EMV coat protein peptides and two gel slices (each 14 cm x 0.75 mm x approx 1.5 mm) cut out. These slices were then separately soaked at 25” in the carboxymethylation solution described by Crestfield et aE. (1963) for 4 hr. For proteolysis, six pieces (each 8 mm wide) of the appropriate gel slice were inserted into the slots of a 20% gel, covered with varying volumes of trypsin-TPCK (Worthington; 2.5 mg/ml of 100 mM CaCl,), and subjected to electrophoresis as described by Cleveland et al. (1977). The gel was PPO-impregnated, stained, and fluorographed as described above.
weight of 2.06 x lo6 was obtained for EMV RNA by sedimentation through linear-log sucrose gradients (data not shown). The RNA was not entirely denatured under conditions used for the sucrose gradient estimation and thus would be expected to yield a high value for the molecular weight. Therefore, we have used 1.9 x 10’ as the molecular weight for EMV RNA in this study. It is possible that the shoulder seen on the gel filtration profile (Fig. 1) resulted from the presence of different conformational forms of this molecule.
RESULTS
In a comparison of the translation properties of TYMV RNA and EMV RNA, Klein (1976) removed spermidine from TYMV RNA with a high salt treatment and also heated the RNA in the presence of EDTA. Although no change in Mg2+ optimum for protein synthesis was observed, for TYMV RNA both treatments resulted in translation of polypeptides of higher molecular weight than for nontreated controls, and also some increase in the amount of coat protein synthesized was seen. Despite the
Column Filtration and Acrylamide Analysis of EMV RNA
Gel
Filtration on acrylamide-agarose gel columns was found to effectively separate tRNA contaminants from the genome RNA of EMV (Pinck and Hall, 1978). Although Ultrogel AcA 22 has an upper exclusion limit of about lo6 MW, we noted that after filtration on a long (136 x 0.8 cm) column, a shoulder could be observed on the lowmolecular-weight side of the genome peak at fractions 19 to 22 (Fig. 1). Attempts to characterize EMV RNA by electrophoresis in agarose-acrylamide gels were not satisfactory, apparently because of the propensity of this RNA to aggregate. However, after denaturation, reliable migration on formamide-containing gels was obtained (Fig. 2). A single RNA band was observed which migrated to a position similar to that of tobacco mosaic virus (TMV) RNA (2 x lo6 daltons; Boedtker, 1968), although a faint trace of a band of roughly 0.5 x lo6 daltons was seen when high levels of EMV RNA were applied to the gels. Alfalfa mosaic virus (AMV) RNAs (Pinck and Hirth, 1972) and brome mosaic virus (BMV) RNAs (Lane and Kaesberg, 19’73)were used as markers to establish a molecular weight of 1.9 x lo6 for EMV RNA (Fig. 2). Although lower than the weight of 2.35 x lo6 cited by Bouley et al. (19’76), this figure is in agreement with the comigration of EMV RNA and TMV RNA seen in Fig. 2. An apparent molecular
Cell-Free Synthesis of Virus Proteins
RNA-Coded
I
FIG. 1. Gel filtration of EMV RNA. RNA extracted from EMV after five cycles of differential centrifugation was applied to a 136 x O.&cm column of AcA 22 gel and eluted with cacodylate buffer. Fractions 16-19 were collected as the heavy-molecular-weight side (Aa) of the genome peak and fractions 20-23 (Ab) as the light side. Peak B contains the contaminating tRNAs.
EMV RNA AMINOACYLATION
.-x
AND TRANSLATION
357
-a-
123456 GEL FIG. 2. Estimation of EMV RNA molecular weight by formamide gel electrophoresis. Electrophoretic migration of viral RNAs in 3.5% polyacrylamide gels containing 99% formamide (anode at bottom) is shown in the left panel. RNAs used were: gel 1, TMV (14 Fg); gel 2, TMV (14 pg) + EMV (16 pg); gel 3, EMV (16 pg); gel 4, EMV (16 pg) + BMV (18.5 pg); gel 5, EMV (32 pg); and gel 6, EMV (16 pg) + AMV (28 pg). To the right, migrations of RNA components seen in gels 4 and 6 are plotted as a function of molecular weight. Using molecular weights (X 10m6)of 1.3, 1.0, 0.7, and 0.34 for the AMV RNA components, and of 1.1, 0.99,0.7, and 0.38 for the BMV RNA components, a molecular weight of 1.9 x lo6 was obtained for the formamide-denatured EMV RNA.
reported lack of effect of EDTA on EMV RNA translation, we also chose to treat some samples with EDTA before column separation and subsequent messenger assay. A comparison of [35S]Met- and [3H]Leulabeled products from incorporation using BMV RNA and EMV RNAs as messengers is shown in Fig. 3. The gel has a 7-14% acrylamide gradient rather than the uniform concentration used by Shih and Kaesberg (19’76). Figure 3 shows a single highmolecular-weight zone for products synthesized using BMV RNA. We assume this to contain the translation products of BMV RNAs 1 and 2, MW 120,000 and 110,000, respectively. Zones corresponding to products of RNA 3 (MW 35,000) and RNA 4 (MW 20,300) also are visible in lanes A, 1, 2, 5, and 6 of Fig. 3. For EMV RNA, a more complex pattern of translation
products, which differed in some respects for [35S]Met- and [3H]Leu-labeled products, was apparent. From lanes 3 and 4 a [35S]Metlabeled polypeptide of apparent MW 21,000 and another at 26,500 were dominant, although bands of apparent MW 30,500, 71,000, and 80,000 are prominent. In contrast, the dominant r3H]Leu-labeled product has an apparent molecular weight of 80,000 (lanes ‘7, 8, and B) although the peptides of approximately MW 21,000 and 26,500 are again present in significant quantity. The reaction shown in lanes 7 and 8 incorporated 28.5% of the supplied [3H]Leu and a 50-fold stimulation over the “minus EMV RNA” control was obtained. Because of the differences between the profiles of [35S]Met- and [3H]Leu-labeled products, a mixture of these isotopes was used in most of the subsequent experiments.
HALL ET AL.
358
A123456
7
8
B
Lane FIG. 3. Fluorograph of translation products from EMV RNA and BMV RNA. After !IO-min incubation, reactions were processed as detailed in the text. The first five lanes show products of [USlMet incorporation; the second five, of [sH]Leu incorporation. Lanes 1 and 2 are of products using BMV RNA as template (lane 1: 4 ~1; lane 2: 8 ~1sample). Lanes 3 and 4 show EMV RNA-directed products (lane 3: 8 pL1;lane 4: 4 ~1). Lanes 5-8 are as for lanes 1-4, but contained 3H-labeled products. Lanes A and B are the same as lanes 1 and 7, but were exposed on less-sensitive film. Radioactivity per 5 ~1 of 50-~1reaction (cpm by disk counting) was: lanes 1 and 2,81,000; lanes 3 and 4,31,000; lanes 5 and 6, 157,ooO;and lanes 7 and 8,190,OOO.Molecular weights from the positions of known marker proteins are shown to the left, and derived maleeulas weights far EMV RNA-coded products are indicated at the right.
Effect of Added Messenger, tRNA, and Chemical Modifiation on the Product Profiles
In addition to the sharp Mg** optimum (3.75 mM) typical of translation with the wheat germ system, we found striking responses to the amounts of tRNA and EMV RNA added to the reactions. For example, incorporation values (cpm&fil sample) of 51,610, 125,430, and 36,050 were obtained on addition of 0, 7.5, and 15 E*g tRNA (wheat germ), respectively (at 3.75 pg EMV RNA per 50-~1 incubation). In the same experiment 3700, 126,430, and 54,825 cpm were obtained for 0, 3.75, and 7.5 rug EMV RNA (in the presence of 7.5 Ng tRNA). Except where otherwise noted, 3.75 n&f MgOAc, was used in all reactions, and no tRNA was added. Rather sharp viral RNA optima have previously been observed for translation of TYMV RNA (Benicourt and Haenni, 1976) and EMV RNA (Klein, 1976). Although the relative effect
of addition of RNA was consistent between experiments, these interacting dependencies, together with the known sensitivity to the levels of other components of the reaction, make compar&ons. of activity between different series of experiments open to question. It was previously postulated that EMV coat protein may be translated in host cells from a small RNA derived from virion RNA by partial replication or specific cleavage (Klein et al., 1976) as may occur for the LMC or TMV RNA (Hunter et al., 1976). More recently, a 250,000-dalton RNA purified from light EMV particles was shown to code for a peptide of 20,000 daltons that comigrated with EMV coat protein (Szybiak et aZ., 1978). A mRNA fraction which was expected to contain the EMV coat protein template was prepared from infected Datura leaves. A peptide product of 22,000 daltons was synthesized in reactions containing this RNA (Fig. 4, lanes 6 and 8), but was only present in trace
EMV RNA AMINOACYLATION
AND TRANSLATION
359
from EMV-infected Datura leaves; BMV RNA-coded products were also included as a control. Electrophoresis of EMV coat protein in i’- 14% gradient gels yielded two polypeptide bands, the more intense having an apparent molecular weight of 21,000 and the other of about 22,000 (lane 4, Fig. 5). The fluorographic pattern of the cell-free products revealed bands coincident with these two polypeptides for reactions coded by EMV RNA (lane 2) or by the mRNA preparation (lane 3, Fig. 5); as expected BMV coat protein synthesized in vitro ran slightly faster than the EMV coat protein marker (lane 1, Fig. 5). The band corre12
3
4
5
6 Lane
7
6
910
1112
FIG. 4. Fluorograph of translation products using different RNA templates. The cell-free protein synthesis reactions contained: lane 1, no added template; lane 2, BMV RNA; lane 3, total EMV RNA; lane 4, EMV RNA fraction Aa (Fig. 1); lane 5, EMV RNA fraction Ab (Fig. 1); lanes 6 and 7, mRNA purified by AcA 22 gel column filtration after extraction from EMV-infected Datum. leaves (lane 6) or healthy leaves (lane 7); lanes 8 and 9, mRNA as in lanes 6 and ‘7 but not passed through a column; lane 10, as lane 3; lanes 11 and 12, EMV RNA after treatment with pancreatic RNase at 0” for 15 min at enzyme:substrate ratios of 1:lOOO(lane 11) or l:lO,OOO (lane 12). Ap proximately equal amounts of radioactivity were added to each lane. The arrows mark the positions of EMV coat protein peptides of MW 22,000 and 20,000.
quantities in products translated using purified genomic RNA as messenger (Fig. 4, lanes 3, 4, 5, and 10). The Z&000-dalton band was more intense when mRNA was used which had not been purified by gel filtration; since only the RNA corresponding to EMV genome RNA, in elution positions 15-18 (Aa of Fig. 1) or 19-22 (Ab of Fig. 1) was tested, a smaller mRNA would tend to have been lost. Szybiak et al. (1978) also noted the occasional synthesis of a 22,000-dalton polypeptide and ascribed its synthesis to the presence of an RNA of approximately 0.65 x lo6 daltons, derived from middle component vu-ions. To ascertain if EMV coat protein was synthesized in the cell-free reactions, a mixture of authentic coat protein was coelectrophoresed with [35S1Met-labeled products of reactions coded by EMV RNA or mRNA
2 2
12345
A6 Lane
FIG. 5. Evidence for coat protein synthesis directed by EMV RNA. Cell-free translation products (separated on a ‘7-148 gradient gel) coded by EMV RNA (lane l), EMV RNA (lane 2) and EMV-infected Datum leaf “mRNA” (lane 3) were identified by fluorography and compared with the position of authentic EMVcoat protein, which was run on the same gel and identified by Coomassie staining (lane 4). Lane 5 contained a reaction sample similar to that run in lane 2 but was stained and reveals the many proteins present in the wheat germ S23 preparation used. The positions of the MW 22,000 and 21,000 polypeptides of EMV coat protein are indicated. The differential synthesis of the EMV coat polypeptides can be more clearly seen in the autoradiograph of products from a separate experiment using EMV RNA as messenger (lane A) or infected Datura leaf “mRNA” (lane B).
HALL ET AL.
360 8:
STAINED
GEL
C: FLUOROGRAPH
FIG. 6. Tryptic digestion of authentic EMV coat protein and EMV RNA-directed translation products. EMV RNA purified by AcA 22 gel filtration and subsequently by sucrose density gradient centrifugation was translated in vitro using [$H]Leu as radioactive substrate. A sample of the labeled products was coelectrophoresed with unlabeled authentic EMV coat protein on a 16% acrylamide gel (Panel A). Two radioactive bands (marked with arrows, Panel A) had mobilities identical to those of the MW 22,000 and 21,000 peptides of the coat protein. These were cut from the gel, subjected to Scarboxymethylation and then to partial tryptic hydrolysis and electrophoretic separation on a 20% gel (see Materials and Methods). Identical profiles were obtained for the stained gel (Panel B) and for the fluorograph (Panel C) of the tryptic peptides. Lane 1 is a control to which no protease was added; differing levels of trypsin were added to the other lanes (25 pg for lanes 2 and 3; 50 pg for lanes 4 and 5; 75 pg for lanes 6 and 7). The trypsin can be seen as a stained band of varying intensity in Pane1 B. The MW 21,000 peptide was added to lanes 1, 2, 4, and 6; the MW 22,000 peptide to lanes 1, 3, 5, and 7.
sponding to 22,000 daltons was more distinct than that at 21,000-daltons for reactions coded by the mRNA, but with EMV RNA as messenger, the 21,000-dalton band (which precisely corresponded with the position of the major polypeptide of EMV coat protein) was predominant. This distinction is even clearer in products of another experiment, shown in lanes A and B, Fig. 5. EMV coat protein was found to be resistant to hydrolysis by chymotrypsin, V8 protease from Staphylococcus aureus, and papain. However, it can be digested with trypsin after S-carboxymethylation (Collot et al., 197’7). A series of partial digestion products was obtained (Fig. 6) using trypsin in the peptide mapping procedure of Cleveland et al. (1977) for both the 22,000- and 21,000-dalton polypeptides. Codigestion of the radioactive peptides obtained by cellfree translation of the EMV RNA with the 22,000- and 21,000-dalton polypeptides of authentic coat protein yielded an identical
profile (compare panels B and C of Fig. 6), confirming that these products do represent viral protein polypeptides. The 22,000dalton band was clearly present on electrophoretic separation of authentic (unlabeled) samples of EMV coat protein (see lane 4 of Fig. 5), and it appears possible that it represents either a second type of coat protein polypeptide or a precursor form of the 21,000-dalton peptide. The 22,000-dalton polypeptide was more resistant to proteolytic digestion than was the 21,000-dalton polypeptide, and although a resemblance in the tryptic pattern could be discerned (by comparison of lanes 2 and 3 of Fig. 6B), we were unable to show conclusively that the 22,000- and 21,000-dalton polypeptides are closely related. In another series of experiments, the effect of added tRNA fractions from infected and healthy Datura leaves and from wheat germ on incorporation was compared. While differences in the efficiency of incorporation
EMV RNA AMINOACYLATION
AND TRANSLATION
361
TABLE 1 EFFECT OF ADDITION OF tRNAs ON THE MESSENGERAmm
OF VUIOUS EMV RNA SAWLES
tRNA added Incorporation (cpm)” Messenger EMV RNA
EMV RNA Aa EMV RNA Ab Dat. mRNA.1
EMV VaI-RNA EMV RNA Aa + Pane RNase IO-’ EMV RNA Aa + Pane RNase lo+
Source
Amount (PLg)
4Omin
lOOmin
None Dat. I Dat. I Dat. H Dat. H WG Dat. I Dat. H WG Dat. I Dat. H WG Dat. I Dat. H WG Dat. I WG Dat. I WG Dat. I WG
0 3 6 3 6 3 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
16,269 22,095 (2) 18,112 35,610 25,941 20,172 (5) 20,119 18,139 14,337 14,484 18,958 18,908 10,364 14,108 12,006 7,299 12,945 9,814 14,042 3,534 2,878
28,523 (1) 31,886 (3) 28,387 53,484 (4) 47,602 33,947 (6,20) 38,060 (9) 37,235 (10) 26,966 (11) 24,857 (12) 38,518 (13) 39,477 (14) 20,245 (15) 29,068 (16) 23,808 (17) 15,540 (7) 29,103 (8) 19,376 (18) 27,435 (19) 5,312 4,060
DAll cpm are for 5~1 samples of the 50-~1 reaction; a mean value of 967 cpm (max 1076, min 848 cpm) for the zero time reaction has been subtracted. Numbers in parentheses denote lane numbers on the product gel shown in Fig. 7. All RNAs were adjusted to 1 mg/ml before addition to the reaction; 4 l~gofeach messenger fraction was used. Each reaction contained 13 ~1 of a 0.5:5.5:3:22 mixture of [35S]Met:[JHJLeu:]l]VaI:[3H]Tyr: [3H]Lys (each 2 mCi/ml) and other components as under Materials and Methods. Pancreatic RNase treatment was at RNA:enzyme ratios of l:lO-’ or l:10m3 (by weight), as indicated. tRNA isolated from EMV-infected (Dat.1) or healthy (Dat.H) Datura leaves, or from wheat germ (WG), and purified by column filtration on AcA 54 Ultrogel, was added to the 50-~1reactions at the concentrations indicated.
were observed (Table l), no major difference in the product profiles was discerned (Fig. ‘7). Donation of Bound Amino Acid by Viral and tRNAs
Previous studies on donation of the amino acid bound to viral RNA during in vitro protein synthesis have led to different conclusions. Chen and Hall (1973) showed that the very low levels of apparent transfer of tyrosine from BMV RNA could be accounted for by small levels of contaminating tRNA. In contrast, Haenni et al. (1973) interpreted their evidence as support for valine donation by TYMV RNA during
protein synthesis. Since EMV is related to TYMV, we thought it would be useful to test the ability of highly purified EMV Val-RNA to donate the bound valine (Table 2). The valylated EMV RNA and tRNA from infected Datum leaves were purified by filtration on an AcA 34 column (P&k and Hall, 1978). Except where the viral RNA was to be tested as both messenger template and valine donor (reactions 15 and 17, Table 2), each reaction contained the same amount of tRNA (either valylated or not) and of valylated or nonaminoacylated) messenger (except in reactions 16, 17, and the template-omitted controls, reactions 1, 10, and 18). Comparison of valine incorporation for reactions 2 and 14 showed
HALL
362
1 2
3
4
5
6
7
6
9
ET AL.
10 11 12 13
14 15
16 17
1619
20
FIG. 7. Translation products coded by various messengers in the presence of different tRNA fractions. Reactions are described in detail in Table 1; lane numbers here correspond to the reaction numbers. Approximately equal amounts of radioactivity were applied to each lane.
that the aminoacylated EMV RNA had the same template activity as did the EMV RNA not subjected to enzymatic aminoacylation. In a separate experiment, translation products using this RNA as template were shown to be similar to those directed by nonaminoacylated RNA (lanes 6 and 8, Fig. 7). The Datura tRNA preparation was an efficient donor, over 50% of the radioactive valine bound being donated using BMV RNA as messenger (reaction 6), and over 25% using EMV RNA as messenger (reaction 8). Kinetics of reaction 8 (Fig. 8, curve c) using Val-tRNA as the isotope source show that it had not gone to completion and longer incubation would have resulted in a higher donation percentage. Although reactions 11 and 14 of Table 2 show that valylated EMV RNA could function as a messenger template, comparison with the data of reactions 13 and 15 reveals that no valine bound to the viral RNA was incorporated in reaction 13 and only 24 cpm in reaction 15. This difference of 24 cpm between the 45- and 0-min samples for reaction 15 calculate out (after allowing for differing counting efficiencies) to represent a 0.7% donation of the added radioactivity (24,144 cpm) from EMV C3H]Val-RNA. However, comparison of the kinetics of this reaction (Fig. 8, curve b) with those of curve a of Fig. 8 (where [3H]Val-tRNA was the valine donor; Table 2, reaction 8) shows no time dependence for the apparent donation from the viral RNA. When low levels of radioactivity (96,500 cpm) were supplied
as free valine (Table 2, reactions l&20), typical incorporation kinetics were obtained (Fig. 8, curve c). Therefore, the 24-cpm difference for reaction 15 of Table 2 should be attributed to experimental variation (equivalent to a pipetting error of +0.5 ~1). A similar rationale holds for the apparent donation of 9 cpm in the presence of BMV RNA (Table 2, reaction 16). Reactions 1, 10, and 18 contained no added messenger template and represent incorporation due to endogenous mRNA in the wheat germ S23. The addition of a large excess of unlabeled valine caused only small decreases in radioactivity donated by tRNA; it is probably significant that in experiments with TYMV Val-RNA, the small amount of apparent donation (Haenni et al., 1973) was drastically reduced by addition of unlabeled amino acid. We conclude that the data shown in Table 2 show no donation of valine from EMV in protein synthesis dependent on EMV RNA or BMV RNA. DISCUSSION
Because we were principally interested in the characteristics of the large genomic RNA, no attempt was made to enrich our preparations in the proportion of smaller RNAs found in the middle and light virions. Indeed, the gel filtration step (Fig. 1) would be expected to eliminate such smaller RNAs. However, because of the tendency of EMV RNA to aggregate, it is difficult to rigorously exclude the presence of viral RNAs smaller than the 1.9 x 106-dalton RNA. Hence, despite the absence of a 0.25
EMV RNA AMINOACYLATION TABLE
AND TRANSLATION
363
2
COMPARISONOF VIRAL AND tRNAs AS VALINE DONORSIN PROTEIN SYNTHESIS Reaction components and radioactivity supplied” Radioactivity incorporated Free amino acid
Transfer RNA
Viral RNA cpm/lO ccl
Reaction Val No. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
+
WI
WI
ValEMV Val Val Dat.1 Dat.1 RNA (9 x 106) (96,500) (0) (85,684) (0)
+ + + + +
+ + + +
+ + + + + +
+
+
+ + +
RNA (0)
+ + + + +
+ + + +
Val-RNA w,w
+ + +
+ +
+ + +
BMV
+ + + + + + + + + +
+
EMV
+ + + + + +
+
+ +
Percent of
45 min 6,463 19,363 989 66,506 73,769 6,562 4,106 3,217 3,354 1,475 3,043 3,879 196 19,354 228 218 198 253 386 888
0 Differmin ence 290 281 278 267 261 191 189 192 263 194 204 212 197 297 204 209 198 204 213 225
6,173 19,082 711 66,239 73,508 6,371 3,917 3,025 3,091 1,281 2,389 3,667 -1 19,057 24 9 0 49 173 663
amount supplied 0.49
1.60 0.06 5.34 5.87 53.91 33.14 25.60 0.25 10.84 18.74 31.03 0.00 1.53 0.72 0.27 0.00 0.37 1.29 4.98
a Other components were as noted under Materials and Methods for protein synthesis; radioactivity supplied is based on solution counting (29% efficiency), while that incorporated on disk counting at 20% efficiency; this difference was taken into account, together with correction to 50-~1total volume in calculating efficiency of incorporation of supplied activity. Amounts of RNA added to each reaction were: 2.5 pg tRNA, 4 Kg EMV RNA, and 6 pg BMV RNA. The unlabeled vahne was at a fmal concentration of 1.2 mM, about 106-foldthe amount of undiluted [3H]Val (26 Wmmol, 2 mCi/ml) supplied as free amino acid. This batch of t3H]Val was also used for aminoacylation of viral RNA and tRNA isolated from infected Datura leaves (Dat.1).
ucts was observed using aminoacylated or x 106-dalton RNA in the electrophoretic profile of Fig. 2, we cannot conclude that the acetylated aminoacylated viral RNA as messenger. viral coat protein peptides are translated The results of Table 2 show conclusively from the large genomic RNA. Some differences in product profile were that the EMV Val-RNA does not function apparent in translation reactions containing as an amino acid donor in protein synthemRNA prepared from purified virus and sis. The only published studies which have suggested the possibility of such a functhat prepared from EMV-infected Datura leaves (Fig. 4). However, our data revealed tion for an aminoaeylated viral RNA (Haenni no differences in polypeptide products et al., 1973) were with less extensively resulting from the addition of tRNA from purified materials and relied on valinol as an healthy or infected Datura leaves or from inhibitor of aminoacylation of (and subsewheat germ, despite their significant ef- quent donation from) tRNA with radioactive valine released from the aminoacylated fect on the total amount of radioactivity incorporated during protein synthesis. viral RNA. Unless some convincing data Similarly, no significant difference in prod- are produced, it would seem that amino-
HALL ET AL.
364
acterization and partial amino acid sequence determination. Proc. Nat. Acad. Sci. USA 73,1964-1968. CHEN, J. M., and HALL, T. C. (1973). Comparison of tyrosyl transfer ribonucleic acid and brome mosaic ,virus tyrosyl ribonueleic acid as amino acid donors in protein synthesis. Biochemistry 12, 45’70-4574. CLEVELAND, D. W., FISCHER, S. G., KIRSCHNER, M. W., and LAEMMLI, II. K. (1977). Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel eleetrophoresis. J. Biol. Chem. 252, 1102-1106. COLLOT, D., PETER, R., PETER, C., and DURANTON, H. (1977). Results relating to the primary structure of the eggplant mosaic virus coat protein. Biochimie
FIG. 8. Kinetics of valine donation. Curve a (0) is taken from reaction 8 of Table 2, which contained Val-tRNA; curve b (0) is from reaction 15, which contained EMV Val-RNA; and curve c(x), from reaction 19, which contained free valine.
acylation of viral RNA has a biological role other than donation in translation processes. ACKNOWLEDGMENTS This work was supported by grants from the NSF (PCM 73-07008), the NIH (AI 11572), the Centre Nationale du Recherche Scientifique (France), and the University of Wisconsin Graduate School and Research Division, College of Agricultural and Life Sciences. REFERENCES BENICOURT, C., and HAENNI, A. L. (1976). In vitro synthesis of turnip yellow mosaic virus coat protein in a wheat germ cell-free system. J. Viral. 20, 196-202. BOEDTKER, H. (1968). Dependence of the sedimentation coefficient on molecular weight of RNA after reaction with formaldehyde. J. Mol. Bill. 35,61-70. BONNER, W. M., and LASKEY, R. A. (1974). A tilm detection method for tritium-labeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biothem. 46,83--88 BOULEY,J. P., BRIAND, J. P., GENEVAUX,M., PINCK, M., and WITZ, J. (1976). The structure of eggplant mosaic virus: Evidence for the presence of low molecular weight RNA in top component. Virology 69, 775-781. CHAN, S. .I., QIM, P., and STEINER, D. F. (1976). Cell-free synthesis of rat preproinsulins: Char-
59, 743-745.
CRESTFIELD, A. M., MOORE, S., and STEIN, W. H. (1963). The preparation and enzymatic hydrolysis of reduced and S-carboxymethylated proteins. J. Biol. Chem. 238, 622-627.
DUESBERG, P. H., and VOGT, P. K. (1973). Gel electrophoresis of avian leukosis and sarcoma viral RNA in formamide: Comparison with other viral and cellular RNA species. J. Viral. 12, 594-599. HAENNI, A. L., PROCHIANTZ,A., BERNARD, O., and CHAPEVILLE, F. (1973). TYMV valyl-RNA as an amino acid donor in protein biosynthesis. Nature New Biol. 241, X6-168. HUNTER, T. R., HUNT, T., KNOWLAND, J., and ZIMMERN, D. (1976). Messenger RNA for the coat protein of tobacco mosaic virus. Nature (London) 260, 759-764. KLEIN, C. (1976). “Variete-structurale du RNA de deux-tymovirus.“Ph.D. thesis, Strasbourg, France. KLEIN, C., FRITSCH, C., BRIAND, J. P., RICHARDS, K. E., JONARD,G., and HIRTH, L. (1976). Physical and functional heterogeneity in TYMV RNA: Evidence for the existence of. an independent messenger coding for coat proteins. Nucleic Acids Res. 3, 3043-3061. KOHL, R. J., and HALL, T. C. (1977). Loss of infectivity of brome mosaic virus RNA after chemical modification of the 3’ or 5’ terminus. Proc. Nat. Acad. Sci. USA 74, 2682-2686.
LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-685. LANE, L. C., and KAESBERG, P. (1973). Multiple genetic components in bromegraas mosaic virus. Nature
New Biol.
232, 40-43.
LAPinor, Y., DE GROOT,N., and FRY-SHAFRIR, I. (1967). Modified aminoacyl-tRNA. II. A general method for the preparation of acylaminoacyl-tRNA. Biochim. Biophys. Acta 145,292-299. MCLEESTER, R. C., and HALL, T. C. (1977). Simplification of amino acid incorporation and other assays using filter paper techniques. Anal. Biochem. 79, 627-630. PINCK, L., and HIRTH, L. (1972). The replicative RNA and the viral RNA synthesis rate in tobacco in-
EMV RNA AMINOACYLATION fected with alfalfa mosaic virus. Virology 49, 413-425. PINCK, M., GHAN, S. K., GENEVAUX, M., HIRTH, L., and DURANTON, H. (1972). Valine specific tRNAlike structure in RNAs of two viruses of turnip yellow mosaic virus group. Biochimie 54, 10931094. PINCK, M., GENEVAUX, M., and DURANTON, H. (1974). Studies on the amino acid acceptor activities of the eggplant mosaic viral RNA and its satellite RNA. Biochimie 56, 423-428. PINCK, M., and HALL, T. C. (1978). Aminoacylation properties of eggplant mosaic virus RNA: Separation and association of tRNAs. Virology 88, 281-285. PLEIJ, C. W. A., MELLEMA, J. R., and BOSCH, L. (1977). The occurrence of the coat protein messenger RNA in the minor components of turnip yellow mosaic virus. FEBS Lett. 80, 19-22. PLEIJ, C. W. A., NEELEMAN, A., VAN VLOTENDOTING, L., and BOSCH, L. (1976). Translation of turnip yellow mosaic virus RNA in vitro: a closed
AND TRANSLATION
365
and an open coat protein cistron. Proc. Nat. Acad. Sci. USA 73,4437-4441. SHIH, D. S., and KAESBERG, P. (1973). Translation of brome mosaic viral RNA in a cell-free system derived from wheat embryo. Proc. Nat. Acad. Sci. USA 70, 1799-1803. SHIH, D. S., and KAESBERG,P. (1976). Translation of RNAs of brome mosaic virus: The monocistronic nature of RNA1 and RNA2. J. Mol. Biol. 103, 7788. STUDIER, F. W. (1973). Analysis of bacteriophage T7 early RNAs and proteins on slab gels. J. Mol. Biol. 79, 237-248. SZYBIAK, U., BOULEY, J. P., and FRITSCH, C. (1978). Evidence for the existence of a coat protein messenger RNA associated with the top component of each of three tymoviruses. Nucleic Acids Res. 5, 1821-1831. WEBER, K., and OSBORN,M. (1969). The reliability of molecular weight determinations by dodecyl sulfate polyacrylamide gel electrophoresis. J. Biol. Chem. 244, 4406-4412.
97, 354-365 (1979)
Aminoacylation
and Messenger
Functions
of Eggplant
Mosaic
Virus RNA
T. C. HALL,*,’ MONIQUE PINCK,? Y. MA, H. M. DURANTON,S AND?‘. L. GERMAN * Department Molkulaire
of Horticulture, University of Wisconsin, Madison, Wisconsin 53706; tlnstitut o!~ Biologie at Cellulaire du CNRS, 15 rue Descartes, 67000-Strasbourg, France; and Slnstitut de Botanique, 28 rue Goethe, 6708%Straebourg, France Accepted May 1, 1979
The predominant RNA of eggplant mosaic virus (EMV) was found to have a molecular weight of 1.9 x lo6 by formamide gel electrophoresis. Electrophoretic analysis of virion protein under dissociating conditions revealed two polypeptides, a major component of 21,000 daltons, and a minor component of 22,000 daltons. Both peptides were present in translation products coded by RNA isolated from virions, but the proportion of the 22,000dalton peptide was higher in products synthesized using RNA isolated from EMV-infected Datura leaves as messenger. Since the viral RNA showed a marked tendency to aggregate, it is possible that these polypeptides were translated from trace amounts of small EMV RNAs present as contaminants of the 1.9 x 105-dalton genomic RNA. Although the addition of tRNA from infected or healthy Datura leaves, or from wheat germ, stimulated amino acid incorporation, no changes were discerned in the profile of cell-free translation products after electrophoretic separation. Nonaminoacylated and valylated EMV RNA stimulated similar levels of amino acid incorporation, and the translation products appeared identical. Valine bound to genomic EMV RNA was not donated during protein synthesis.
the 5’-end and in that the 3’-terminus acUntil recently, the tymoviruses, includ- cepts valine (Klein, 1976; Pinck et al., 1972). ing eggplant mosaic virus (EMV) and turnip Additionally, EMV RNA typically has subyellow mosaic virus (TYMV), were thought stantial quantities of associated or comto have only a single RNA component. Klein plexed tRNA (Pinck et al., 1974; Bouley et al. (1976) have described the structure et al., 1976; Pinck and Hall, 1978). In this and messenger function of two distinct study we examined the influence of added RNA components of TYMV of 2 x lo6 and tRNA and aminoacylation on translation of 0.3 x lo6 daltons, and Pleij et al. (1976) EMV RNA. We also investigated if any donation of valine from the 3’-end of the found that mild heat treatment of TYMV genome RNA occurred during protein synRNA yielded a series of RNA classes, the thesis. smallest of which was an efficient messenger for coat protein. Coat protein was also MATERIALS AND METHODS observed among the translation products Virus and RNA preparation. EMV and using unfractionated TYMV RNA as mesEMV RNA were prepared as described by senger (Benicourt and Haenni, 1976). Pinck and Hall (1978), LKB Ultrogel AcA Bouley et al. (1976) and Szybiak et al. (1978) showed that three classes of virions 22 being typically used for purification of could be separated by equilibrium cen- RNA. tRNA was extracted according to the trifugation of EMV in CsCl. Heavy (bot- procedure of Chan et al. (1976) from healthy tom component) virions predominate and and EMV-infected Datura leaves and from contain infective RNA of about 2 x lo6 wheat germ. Much of the contaminating daltons. EMV RNA is similar to TYMV DNA was removed by spooling onto a glass RNA in having an m7GpppX structure at rod; the remainder was eliminated during gel filtration on a 130 x 0.7-cm column of ’ To whom requests for reprints should be sent. AcA 54 gel. The RNA insoluble in 2 M NalNTRODUCTION
0042-6822/79/120354-12$02.00/O Copyright All rights
0 19’79 by Academic Press, Inc. of reproduction in any form reserved.
354
EMV RNA AMINOACYLATION
acetate (obtained during tRNA preparation from total nucleic acid extracts of healthy leaves) was dissolved and infected htura in water, precipitated with ethanol, redissolved in water, and filtered through a 72 x 0.7~cm column of AcA 34 gel using 10 mM Na-cacodylate buffer (pH 5.7) containing 0.1 M NaCl and 3 mM NaN,. Nucleic acid eluting as a peak of excluded material was precipitated with ethanol and subsequently used as a crude messenger (mRNA) fraction. RNA eluting at other positions showed no messenger activity in vitro. For electrophoresis, EMV RNA was dissolved in 4 M urea and an equal volume of formamide added. After heating to 60” for 5 min, the samples were quench-cooled in ice water and then applied to 3.5% polyacrylamide gels (0.6 x 10 cm) containing 99% formamide (Duesberg and Vogt, 1973). Electrophoresis at 80 V was carried out at room temperature for 17 hr. Gels were stained in 0.1% (w/v) toluidine blue in 1% acetic acid for 1 hr. Synthetase and aminoacylation reactions. These were as described by Pinck
and Hall (1978), except that the reaction volumes were scaled up to as much as 1 ml for preparative aminoacylation of tRNAs and EMV RNA fractions. Acetylation of aminoacylated RNA was according to Lapidot et al. (1967). Aminoacylated and acetylated RNA derivatives were purified by filtration (acidic buffer containing 0.1 M NaCl and 3 mM NaN, in 10 mM-Na cacodylate, pH 5.7, being used for all steps) on AcA 54 gel (for tRNAs) or AcA 34 gel (for viral RNAs), followed by precipitation with ethanol. Amino acid donation experiments were similar to those of Chen and Hall (1973). Cell-free protein synthesis and product electrophoresis. Wheat germ S23 prepara-
tion and amino acid incorporation conditions were as described by Kohl and Hall (1977), except that a French press was used to disrupt the wheat germ and [35S]Met, [3H]Leu, and [3H]Val were used as radioactive substrates. Incorporation was measured by taking 5-~1 samples of the 50-~1 reactions for assay according to the filter paper disk technique of McLeester and Hall
AND TRANSLATION
355
(1977), and reactions were stopped by freezing and storage at -20”. For product analysis, portions of the reactions were dissociated by heating (85” for 6 mm) with an equal volume of a solution (pH 6.8 with HCl) containing 10 mil! Tris, 10 mM EDTA, 100 mil4 NaCl, 4% (w/v) mercaptoethanol, and 20% (v/v) glycerol. A trace of bromophenol blue was added as tracking dye. Sample volumes (2 to 12 ~1) were chosen to contain about 50,000 cpm (by the disk assay) and applied to Laemmli (1970) gels (containing a 7-14% acrylamide gradient) and subjected to electrophoresis (anode at the bottom) in a Studier (1973) apparatus. For fluorography (Bonner and Laskey, 1974), these 0.75-mm-thick gels were soaked in two changes of DMSO (15 min each), then covered with DMSO containing 20% (v/v) PPO for 30 min, and shaken gently. On placing in water, the gel expanded to its original size within 2 min and was matched with a slice taken from the same gel containing EMV coat protein (a kind gift of Dr. D. Chollot, Strasbourg) and other standard proteins (Boehringer-Mannheim “Combithek” molecular weight standards: trypsin inhibitor, 21,500; bovine serum albumin, 68,000; Escherichia coli RNA polymerase a, 39,000; /3, 155,000; p’, 165,000) and stained with Coomassie brilliant blue (Weber and Osborn, 1969). The gel was dried onto filter paper (Studier, 1973), then exposed to Kodak No-Screen or Royal X-omat X-ray film. In later experiments, the PPO-impregnated gel was swollen in water, then stained for 3 min with a fresh solution of Coomassie brilliant blue and destained prior to drying. This procedure only slightly detracted from the efficiency of fluorography and permitted direct comparison of the migration of stained markers (mixed with reaction products) and the radioactive products. Cell-free reaction products coded by BMV RNA were also used as radioactive markers. Peptide mapping. A mixture (300 ~1) of authentic EMV coat protein (Collot et al., 1977) and [3H]Leu-labeled EMV RNAcoded translation products (1.7 x lo6 dpm) were dissociated as described above and applied to a 16% Laemmli (1970) gel. After electrophoresis, the gel was briefly stained (Weber and Osborn, 1969) to locate the
HALL ET AL.
356
22,000 and 21,000 MW EMV coat protein peptides and two gel slices (each 14 cm x 0.75 mm x approx 1.5 mm) cut out. These slices were then separately soaked at 25” in the carboxymethylation solution described by Crestfield et aE. (1963) for 4 hr. For proteolysis, six pieces (each 8 mm wide) of the appropriate gel slice were inserted into the slots of a 20% gel, covered with varying volumes of trypsin-TPCK (Worthington; 2.5 mg/ml of 100 mM CaCl,), and subjected to electrophoresis as described by Cleveland et al. (1977). The gel was PPO-impregnated, stained, and fluorographed as described above.
weight of 2.06 x lo6 was obtained for EMV RNA by sedimentation through linear-log sucrose gradients (data not shown). The RNA was not entirely denatured under conditions used for the sucrose gradient estimation and thus would be expected to yield a high value for the molecular weight. Therefore, we have used 1.9 x 10’ as the molecular weight for EMV RNA in this study. It is possible that the shoulder seen on the gel filtration profile (Fig. 1) resulted from the presence of different conformational forms of this molecule.
RESULTS
In a comparison of the translation properties of TYMV RNA and EMV RNA, Klein (1976) removed spermidine from TYMV RNA with a high salt treatment and also heated the RNA in the presence of EDTA. Although no change in Mg2+ optimum for protein synthesis was observed, for TYMV RNA both treatments resulted in translation of polypeptides of higher molecular weight than for nontreated controls, and also some increase in the amount of coat protein synthesized was seen. Despite the
Column Filtration and Acrylamide Analysis of EMV RNA
Gel
Filtration on acrylamide-agarose gel columns was found to effectively separate tRNA contaminants from the genome RNA of EMV (Pinck and Hall, 1978). Although Ultrogel AcA 22 has an upper exclusion limit of about lo6 MW, we noted that after filtration on a long (136 x 0.8 cm) column, a shoulder could be observed on the lowmolecular-weight side of the genome peak at fractions 19 to 22 (Fig. 1). Attempts to characterize EMV RNA by electrophoresis in agarose-acrylamide gels were not satisfactory, apparently because of the propensity of this RNA to aggregate. However, after denaturation, reliable migration on formamide-containing gels was obtained (Fig. 2). A single RNA band was observed which migrated to a position similar to that of tobacco mosaic virus (TMV) RNA (2 x lo6 daltons; Boedtker, 1968), although a faint trace of a band of roughly 0.5 x lo6 daltons was seen when high levels of EMV RNA were applied to the gels. Alfalfa mosaic virus (AMV) RNAs (Pinck and Hirth, 1972) and brome mosaic virus (BMV) RNAs (Lane and Kaesberg, 19’73)were used as markers to establish a molecular weight of 1.9 x lo6 for EMV RNA (Fig. 2). Although lower than the weight of 2.35 x lo6 cited by Bouley et al. (19’76), this figure is in agreement with the comigration of EMV RNA and TMV RNA seen in Fig. 2. An apparent molecular
Cell-Free Synthesis of Virus Proteins
RNA-Coded
I
FIG. 1. Gel filtration of EMV RNA. RNA extracted from EMV after five cycles of differential centrifugation was applied to a 136 x O.&cm column of AcA 22 gel and eluted with cacodylate buffer. Fractions 16-19 were collected as the heavy-molecular-weight side (Aa) of the genome peak and fractions 20-23 (Ab) as the light side. Peak B contains the contaminating tRNAs.
EMV RNA AMINOACYLATION
.-x
AND TRANSLATION
357
-a-
123456 GEL FIG. 2. Estimation of EMV RNA molecular weight by formamide gel electrophoresis. Electrophoretic migration of viral RNAs in 3.5% polyacrylamide gels containing 99% formamide (anode at bottom) is shown in the left panel. RNAs used were: gel 1, TMV (14 Fg); gel 2, TMV (14 pg) + EMV (16 pg); gel 3, EMV (16 pg); gel 4, EMV (16 pg) + BMV (18.5 pg); gel 5, EMV (32 pg); and gel 6, EMV (16 pg) + AMV (28 pg). To the right, migrations of RNA components seen in gels 4 and 6 are plotted as a function of molecular weight. Using molecular weights (X 10m6)of 1.3, 1.0, 0.7, and 0.34 for the AMV RNA components, and of 1.1, 0.99,0.7, and 0.38 for the BMV RNA components, a molecular weight of 1.9 x lo6 was obtained for the formamide-denatured EMV RNA.
reported lack of effect of EDTA on EMV RNA translation, we also chose to treat some samples with EDTA before column separation and subsequent messenger assay. A comparison of [35S]Met- and [3H]Leulabeled products from incorporation using BMV RNA and EMV RNAs as messengers is shown in Fig. 3. The gel has a 7-14% acrylamide gradient rather than the uniform concentration used by Shih and Kaesberg (19’76). Figure 3 shows a single highmolecular-weight zone for products synthesized using BMV RNA. We assume this to contain the translation products of BMV RNAs 1 and 2, MW 120,000 and 110,000, respectively. Zones corresponding to products of RNA 3 (MW 35,000) and RNA 4 (MW 20,300) also are visible in lanes A, 1, 2, 5, and 6 of Fig. 3. For EMV RNA, a more complex pattern of translation
products, which differed in some respects for [35S]Met- and [3H]Leu-labeled products, was apparent. From lanes 3 and 4 a [35S]Metlabeled polypeptide of apparent MW 21,000 and another at 26,500 were dominant, although bands of apparent MW 30,500, 71,000, and 80,000 are prominent. In contrast, the dominant r3H]Leu-labeled product has an apparent molecular weight of 80,000 (lanes ‘7, 8, and B) although the peptides of approximately MW 21,000 and 26,500 are again present in significant quantity. The reaction shown in lanes 7 and 8 incorporated 28.5% of the supplied [3H]Leu and a 50-fold stimulation over the “minus EMV RNA” control was obtained. Because of the differences between the profiles of [35S]Met- and [3H]Leu-labeled products, a mixture of these isotopes was used in most of the subsequent experiments.
HALL ET AL.
358
A123456
7
8
B
Lane FIG. 3. Fluorograph of translation products from EMV RNA and BMV RNA. After !IO-min incubation, reactions were processed as detailed in the text. The first five lanes show products of [USlMet incorporation; the second five, of [sH]Leu incorporation. Lanes 1 and 2 are of products using BMV RNA as template (lane 1: 4 ~1; lane 2: 8 ~1sample). Lanes 3 and 4 show EMV RNA-directed products (lane 3: 8 pL1;lane 4: 4 ~1). Lanes 5-8 are as for lanes 1-4, but contained 3H-labeled products. Lanes A and B are the same as lanes 1 and 7, but were exposed on less-sensitive film. Radioactivity per 5 ~1 of 50-~1reaction (cpm by disk counting) was: lanes 1 and 2,81,000; lanes 3 and 4,31,000; lanes 5 and 6, 157,ooO;and lanes 7 and 8,190,OOO.Molecular weights from the positions of known marker proteins are shown to the left, and derived maleeulas weights far EMV RNA-coded products are indicated at the right.
Effect of Added Messenger, tRNA, and Chemical Modifiation on the Product Profiles
In addition to the sharp Mg** optimum (3.75 mM) typical of translation with the wheat germ system, we found striking responses to the amounts of tRNA and EMV RNA added to the reactions. For example, incorporation values (cpm&fil sample) of 51,610, 125,430, and 36,050 were obtained on addition of 0, 7.5, and 15 E*g tRNA (wheat germ), respectively (at 3.75 pg EMV RNA per 50-~1 incubation). In the same experiment 3700, 126,430, and 54,825 cpm were obtained for 0, 3.75, and 7.5 rug EMV RNA (in the presence of 7.5 Ng tRNA). Except where otherwise noted, 3.75 n&f MgOAc, was used in all reactions, and no tRNA was added. Rather sharp viral RNA optima have previously been observed for translation of TYMV RNA (Benicourt and Haenni, 1976) and EMV RNA (Klein, 1976). Although the relative effect
of addition of RNA was consistent between experiments, these interacting dependencies, together with the known sensitivity to the levels of other components of the reaction, make compar&ons. of activity between different series of experiments open to question. It was previously postulated that EMV coat protein may be translated in host cells from a small RNA derived from virion RNA by partial replication or specific cleavage (Klein et al., 1976) as may occur for the LMC or TMV RNA (Hunter et al., 1976). More recently, a 250,000-dalton RNA purified from light EMV particles was shown to code for a peptide of 20,000 daltons that comigrated with EMV coat protein (Szybiak et aZ., 1978). A mRNA fraction which was expected to contain the EMV coat protein template was prepared from infected Datura leaves. A peptide product of 22,000 daltons was synthesized in reactions containing this RNA (Fig. 4, lanes 6 and 8), but was only present in trace
EMV RNA AMINOACYLATION
AND TRANSLATION
359
from EMV-infected Datura leaves; BMV RNA-coded products were also included as a control. Electrophoresis of EMV coat protein in i’- 14% gradient gels yielded two polypeptide bands, the more intense having an apparent molecular weight of 21,000 and the other of about 22,000 (lane 4, Fig. 5). The fluorographic pattern of the cell-free products revealed bands coincident with these two polypeptides for reactions coded by EMV RNA (lane 2) or by the mRNA preparation (lane 3, Fig. 5); as expected BMV coat protein synthesized in vitro ran slightly faster than the EMV coat protein marker (lane 1, Fig. 5). The band corre12
3
4
5
6 Lane
7
6
910
1112
FIG. 4. Fluorograph of translation products using different RNA templates. The cell-free protein synthesis reactions contained: lane 1, no added template; lane 2, BMV RNA; lane 3, total EMV RNA; lane 4, EMV RNA fraction Aa (Fig. 1); lane 5, EMV RNA fraction Ab (Fig. 1); lanes 6 and 7, mRNA purified by AcA 22 gel column filtration after extraction from EMV-infected Datum. leaves (lane 6) or healthy leaves (lane 7); lanes 8 and 9, mRNA as in lanes 6 and ‘7 but not passed through a column; lane 10, as lane 3; lanes 11 and 12, EMV RNA after treatment with pancreatic RNase at 0” for 15 min at enzyme:substrate ratios of 1:lOOO(lane 11) or l:lO,OOO (lane 12). Ap proximately equal amounts of radioactivity were added to each lane. The arrows mark the positions of EMV coat protein peptides of MW 22,000 and 20,000.
quantities in products translated using purified genomic RNA as messenger (Fig. 4, lanes 3, 4, 5, and 10). The Z&000-dalton band was more intense when mRNA was used which had not been purified by gel filtration; since only the RNA corresponding to EMV genome RNA, in elution positions 15-18 (Aa of Fig. 1) or 19-22 (Ab of Fig. 1) was tested, a smaller mRNA would tend to have been lost. Szybiak et al. (1978) also noted the occasional synthesis of a 22,000-dalton polypeptide and ascribed its synthesis to the presence of an RNA of approximately 0.65 x lo6 daltons, derived from middle component vu-ions. To ascertain if EMV coat protein was synthesized in the cell-free reactions, a mixture of authentic coat protein was coelectrophoresed with [35S1Met-labeled products of reactions coded by EMV RNA or mRNA
2 2
12345
A6 Lane
FIG. 5. Evidence for coat protein synthesis directed by EMV RNA. Cell-free translation products (separated on a ‘7-148 gradient gel) coded by EMV RNA (lane l), EMV RNA (lane 2) and EMV-infected Datum leaf “mRNA” (lane 3) were identified by fluorography and compared with the position of authentic EMVcoat protein, which was run on the same gel and identified by Coomassie staining (lane 4). Lane 5 contained a reaction sample similar to that run in lane 2 but was stained and reveals the many proteins present in the wheat germ S23 preparation used. The positions of the MW 22,000 and 21,000 polypeptides of EMV coat protein are indicated. The differential synthesis of the EMV coat polypeptides can be more clearly seen in the autoradiograph of products from a separate experiment using EMV RNA as messenger (lane A) or infected Datura leaf “mRNA” (lane B).
HALL ET AL.
360 8:
STAINED
GEL
C: FLUOROGRAPH
FIG. 6. Tryptic digestion of authentic EMV coat protein and EMV RNA-directed translation products. EMV RNA purified by AcA 22 gel filtration and subsequently by sucrose density gradient centrifugation was translated in vitro using [$H]Leu as radioactive substrate. A sample of the labeled products was coelectrophoresed with unlabeled authentic EMV coat protein on a 16% acrylamide gel (Panel A). Two radioactive bands (marked with arrows, Panel A) had mobilities identical to those of the MW 22,000 and 21,000 peptides of the coat protein. These were cut from the gel, subjected to Scarboxymethylation and then to partial tryptic hydrolysis and electrophoretic separation on a 20% gel (see Materials and Methods). Identical profiles were obtained for the stained gel (Panel B) and for the fluorograph (Panel C) of the tryptic peptides. Lane 1 is a control to which no protease was added; differing levels of trypsin were added to the other lanes (25 pg for lanes 2 and 3; 50 pg for lanes 4 and 5; 75 pg for lanes 6 and 7). The trypsin can be seen as a stained band of varying intensity in Pane1 B. The MW 21,000 peptide was added to lanes 1, 2, 4, and 6; the MW 22,000 peptide to lanes 1, 3, 5, and 7.
sponding to 22,000 daltons was more distinct than that at 21,000-daltons for reactions coded by the mRNA, but with EMV RNA as messenger, the 21,000-dalton band (which precisely corresponded with the position of the major polypeptide of EMV coat protein) was predominant. This distinction is even clearer in products of another experiment, shown in lanes A and B, Fig. 5. EMV coat protein was found to be resistant to hydrolysis by chymotrypsin, V8 protease from Staphylococcus aureus, and papain. However, it can be digested with trypsin after S-carboxymethylation (Collot et al., 197’7). A series of partial digestion products was obtained (Fig. 6) using trypsin in the peptide mapping procedure of Cleveland et al. (1977) for both the 22,000- and 21,000-dalton polypeptides. Codigestion of the radioactive peptides obtained by cellfree translation of the EMV RNA with the 22,000- and 21,000-dalton polypeptides of authentic coat protein yielded an identical
profile (compare panels B and C of Fig. 6), confirming that these products do represent viral protein polypeptides. The 22,000dalton band was clearly present on electrophoretic separation of authentic (unlabeled) samples of EMV coat protein (see lane 4 of Fig. 5), and it appears possible that it represents either a second type of coat protein polypeptide or a precursor form of the 21,000-dalton peptide. The 22,000-dalton polypeptide was more resistant to proteolytic digestion than was the 21,000-dalton polypeptide, and although a resemblance in the tryptic pattern could be discerned (by comparison of lanes 2 and 3 of Fig. 6B), we were unable to show conclusively that the 22,000- and 21,000-dalton polypeptides are closely related. In another series of experiments, the effect of added tRNA fractions from infected and healthy Datura leaves and from wheat germ on incorporation was compared. While differences in the efficiency of incorporation
EMV RNA AMINOACYLATION
AND TRANSLATION
361
TABLE 1 EFFECT OF ADDITION OF tRNAs ON THE MESSENGERAmm
OF VUIOUS EMV RNA SAWLES
tRNA added Incorporation (cpm)” Messenger EMV RNA
EMV RNA Aa EMV RNA Ab Dat. mRNA.1
EMV VaI-RNA EMV RNA Aa + Pane RNase IO-’ EMV RNA Aa + Pane RNase lo+
Source
Amount (PLg)
4Omin
lOOmin
None Dat. I Dat. I Dat. H Dat. H WG Dat. I Dat. H WG Dat. I Dat. H WG Dat. I Dat. H WG Dat. I WG Dat. I WG Dat. I WG
0 3 6 3 6 3 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
16,269 22,095 (2) 18,112 35,610 25,941 20,172 (5) 20,119 18,139 14,337 14,484 18,958 18,908 10,364 14,108 12,006 7,299 12,945 9,814 14,042 3,534 2,878
28,523 (1) 31,886 (3) 28,387 53,484 (4) 47,602 33,947 (6,20) 38,060 (9) 37,235 (10) 26,966 (11) 24,857 (12) 38,518 (13) 39,477 (14) 20,245 (15) 29,068 (16) 23,808 (17) 15,540 (7) 29,103 (8) 19,376 (18) 27,435 (19) 5,312 4,060
DAll cpm are for 5~1 samples of the 50-~1 reaction; a mean value of 967 cpm (max 1076, min 848 cpm) for the zero time reaction has been subtracted. Numbers in parentheses denote lane numbers on the product gel shown in Fig. 7. All RNAs were adjusted to 1 mg/ml before addition to the reaction; 4 l~gofeach messenger fraction was used. Each reaction contained 13 ~1 of a 0.5:5.5:3:22 mixture of [35S]Met:[JHJLeu:]l]VaI:[3H]Tyr: [3H]Lys (each 2 mCi/ml) and other components as under Materials and Methods. Pancreatic RNase treatment was at RNA:enzyme ratios of l:lO-’ or l:10m3 (by weight), as indicated. tRNA isolated from EMV-infected (Dat.1) or healthy (Dat.H) Datura leaves, or from wheat germ (WG), and purified by column filtration on AcA 54 Ultrogel, was added to the 50-~1reactions at the concentrations indicated.
were observed (Table l), no major difference in the product profiles was discerned (Fig. ‘7). Donation of Bound Amino Acid by Viral and tRNAs
Previous studies on donation of the amino acid bound to viral RNA during in vitro protein synthesis have led to different conclusions. Chen and Hall (1973) showed that the very low levels of apparent transfer of tyrosine from BMV RNA could be accounted for by small levels of contaminating tRNA. In contrast, Haenni et al. (1973) interpreted their evidence as support for valine donation by TYMV RNA during
protein synthesis. Since EMV is related to TYMV, we thought it would be useful to test the ability of highly purified EMV Val-RNA to donate the bound valine (Table 2). The valylated EMV RNA and tRNA from infected Datum leaves were purified by filtration on an AcA 34 column (P&k and Hall, 1978). Except where the viral RNA was to be tested as both messenger template and valine donor (reactions 15 and 17, Table 2), each reaction contained the same amount of tRNA (either valylated or not) and of valylated or nonaminoacylated) messenger (except in reactions 16, 17, and the template-omitted controls, reactions 1, 10, and 18). Comparison of valine incorporation for reactions 2 and 14 showed
HALL
362
1 2
3
4
5
6
7
6
9
ET AL.
10 11 12 13
14 15
16 17
1619
20
FIG. 7. Translation products coded by various messengers in the presence of different tRNA fractions. Reactions are described in detail in Table 1; lane numbers here correspond to the reaction numbers. Approximately equal amounts of radioactivity were applied to each lane.
that the aminoacylated EMV RNA had the same template activity as did the EMV RNA not subjected to enzymatic aminoacylation. In a separate experiment, translation products using this RNA as template were shown to be similar to those directed by nonaminoacylated RNA (lanes 6 and 8, Fig. 7). The Datura tRNA preparation was an efficient donor, over 50% of the radioactive valine bound being donated using BMV RNA as messenger (reaction 6), and over 25% using EMV RNA as messenger (reaction 8). Kinetics of reaction 8 (Fig. 8, curve c) using Val-tRNA as the isotope source show that it had not gone to completion and longer incubation would have resulted in a higher donation percentage. Although reactions 11 and 14 of Table 2 show that valylated EMV RNA could function as a messenger template, comparison with the data of reactions 13 and 15 reveals that no valine bound to the viral RNA was incorporated in reaction 13 and only 24 cpm in reaction 15. This difference of 24 cpm between the 45- and 0-min samples for reaction 15 calculate out (after allowing for differing counting efficiencies) to represent a 0.7% donation of the added radioactivity (24,144 cpm) from EMV C3H]Val-RNA. However, comparison of the kinetics of this reaction (Fig. 8, curve b) with those of curve a of Fig. 8 (where [3H]Val-tRNA was the valine donor; Table 2, reaction 8) shows no time dependence for the apparent donation from the viral RNA. When low levels of radioactivity (96,500 cpm) were supplied
as free valine (Table 2, reactions l&20), typical incorporation kinetics were obtained (Fig. 8, curve c). Therefore, the 24-cpm difference for reaction 15 of Table 2 should be attributed to experimental variation (equivalent to a pipetting error of +0.5 ~1). A similar rationale holds for the apparent donation of 9 cpm in the presence of BMV RNA (Table 2, reaction 16). Reactions 1, 10, and 18 contained no added messenger template and represent incorporation due to endogenous mRNA in the wheat germ S23. The addition of a large excess of unlabeled valine caused only small decreases in radioactivity donated by tRNA; it is probably significant that in experiments with TYMV Val-RNA, the small amount of apparent donation (Haenni et al., 1973) was drastically reduced by addition of unlabeled amino acid. We conclude that the data shown in Table 2 show no donation of valine from EMV in protein synthesis dependent on EMV RNA or BMV RNA. DISCUSSION
Because we were principally interested in the characteristics of the large genomic RNA, no attempt was made to enrich our preparations in the proportion of smaller RNAs found in the middle and light virions. Indeed, the gel filtration step (Fig. 1) would be expected to eliminate such smaller RNAs. However, because of the tendency of EMV RNA to aggregate, it is difficult to rigorously exclude the presence of viral RNAs smaller than the 1.9 x 106-dalton RNA. Hence, despite the absence of a 0.25
EMV RNA AMINOACYLATION TABLE
AND TRANSLATION
363
2
COMPARISONOF VIRAL AND tRNAs AS VALINE DONORSIN PROTEIN SYNTHESIS Reaction components and radioactivity supplied” Radioactivity incorporated Free amino acid
Transfer RNA
Viral RNA cpm/lO ccl
Reaction Val No. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
+
WI
WI
ValEMV Val Val Dat.1 Dat.1 RNA (9 x 106) (96,500) (0) (85,684) (0)
+ + + + +
+ + + +
+ + + + + +
+
+
+ + +
RNA (0)
+ + + + +
+ + + +
Val-RNA w,w
+ + +
+ +
+ + +
BMV
+ + + + + + + + + +
+
EMV
+ + + + + +
+
+ +
Percent of
45 min 6,463 19,363 989 66,506 73,769 6,562 4,106 3,217 3,354 1,475 3,043 3,879 196 19,354 228 218 198 253 386 888
0 Differmin ence 290 281 278 267 261 191 189 192 263 194 204 212 197 297 204 209 198 204 213 225
6,173 19,082 711 66,239 73,508 6,371 3,917 3,025 3,091 1,281 2,389 3,667 -1 19,057 24 9 0 49 173 663
amount supplied 0.49
1.60 0.06 5.34 5.87 53.91 33.14 25.60 0.25 10.84 18.74 31.03 0.00 1.53 0.72 0.27 0.00 0.37 1.29 4.98
a Other components were as noted under Materials and Methods for protein synthesis; radioactivity supplied is based on solution counting (29% efficiency), while that incorporated on disk counting at 20% efficiency; this difference was taken into account, together with correction to 50-~1total volume in calculating efficiency of incorporation of supplied activity. Amounts of RNA added to each reaction were: 2.5 pg tRNA, 4 Kg EMV RNA, and 6 pg BMV RNA. The unlabeled vahne was at a fmal concentration of 1.2 mM, about 106-foldthe amount of undiluted [3H]Val (26 Wmmol, 2 mCi/ml) supplied as free amino acid. This batch of t3H]Val was also used for aminoacylation of viral RNA and tRNA isolated from infected Datura leaves (Dat.1).
ucts was observed using aminoacylated or x 106-dalton RNA in the electrophoretic profile of Fig. 2, we cannot conclude that the acetylated aminoacylated viral RNA as messenger. viral coat protein peptides are translated The results of Table 2 show conclusively from the large genomic RNA. Some differences in product profile were that the EMV Val-RNA does not function apparent in translation reactions containing as an amino acid donor in protein synthemRNA prepared from purified virus and sis. The only published studies which have suggested the possibility of such a functhat prepared from EMV-infected Datura leaves (Fig. 4). However, our data revealed tion for an aminoaeylated viral RNA (Haenni no differences in polypeptide products et al., 1973) were with less extensively resulting from the addition of tRNA from purified materials and relied on valinol as an healthy or infected Datura leaves or from inhibitor of aminoacylation of (and subsewheat germ, despite their significant ef- quent donation from) tRNA with radioactive valine released from the aminoacylated fect on the total amount of radioactivity incorporated during protein synthesis. viral RNA. Unless some convincing data Similarly, no significant difference in prod- are produced, it would seem that amino-
HALL ET AL.
364
acterization and partial amino acid sequence determination. Proc. Nat. Acad. Sci. USA 73,1964-1968. CHEN, J. M., and HALL, T. C. (1973). Comparison of tyrosyl transfer ribonucleic acid and brome mosaic ,virus tyrosyl ribonueleic acid as amino acid donors in protein synthesis. Biochemistry 12, 45’70-4574. CLEVELAND, D. W., FISCHER, S. G., KIRSCHNER, M. W., and LAEMMLI, II. K. (1977). Peptide mapping by limited proteolysis in sodium dodecyl sulfate and analysis by gel eleetrophoresis. J. Biol. Chem. 252, 1102-1106. COLLOT, D., PETER, R., PETER, C., and DURANTON, H. (1977). Results relating to the primary structure of the eggplant mosaic virus coat protein. Biochimie
FIG. 8. Kinetics of valine donation. Curve a (0) is taken from reaction 8 of Table 2, which contained Val-tRNA; curve b (0) is from reaction 15, which contained EMV Val-RNA; and curve c(x), from reaction 19, which contained free valine.
acylation of viral RNA has a biological role other than donation in translation processes. ACKNOWLEDGMENTS This work was supported by grants from the NSF (PCM 73-07008), the NIH (AI 11572), the Centre Nationale du Recherche Scientifique (France), and the University of Wisconsin Graduate School and Research Division, College of Agricultural and Life Sciences. REFERENCES BENICOURT, C., and HAENNI, A. L. (1976). In vitro synthesis of turnip yellow mosaic virus coat protein in a wheat germ cell-free system. J. Viral. 20, 196-202. BOEDTKER, H. (1968). Dependence of the sedimentation coefficient on molecular weight of RNA after reaction with formaldehyde. J. Mol. Bill. 35,61-70. BONNER, W. M., and LASKEY, R. A. (1974). A tilm detection method for tritium-labeled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biothem. 46,83--88 BOULEY,J. P., BRIAND, J. P., GENEVAUX,M., PINCK, M., and WITZ, J. (1976). The structure of eggplant mosaic virus: Evidence for the presence of low molecular weight RNA in top component. Virology 69, 775-781. CHAN, S. .I., QIM, P., and STEINER, D. F. (1976). Cell-free synthesis of rat preproinsulins: Char-
59, 743-745.
CRESTFIELD, A. M., MOORE, S., and STEIN, W. H. (1963). The preparation and enzymatic hydrolysis of reduced and S-carboxymethylated proteins. J. Biol. Chem. 238, 622-627.
DUESBERG, P. H., and VOGT, P. K. (1973). Gel electrophoresis of avian leukosis and sarcoma viral RNA in formamide: Comparison with other viral and cellular RNA species. J. Viral. 12, 594-599. HAENNI, A. L., PROCHIANTZ,A., BERNARD, O., and CHAPEVILLE, F. (1973). TYMV valyl-RNA as an amino acid donor in protein biosynthesis. Nature New Biol. 241, X6-168. HUNTER, T. R., HUNT, T., KNOWLAND, J., and ZIMMERN, D. (1976). Messenger RNA for the coat protein of tobacco mosaic virus. Nature (London) 260, 759-764. KLEIN, C. (1976). “Variete-structurale du RNA de deux-tymovirus.“Ph.D. thesis, Strasbourg, France. KLEIN, C., FRITSCH, C., BRIAND, J. P., RICHARDS, K. E., JONARD,G., and HIRTH, L. (1976). Physical and functional heterogeneity in TYMV RNA: Evidence for the existence of. an independent messenger coding for coat proteins. Nucleic Acids Res. 3, 3043-3061. KOHL, R. J., and HALL, T. C. (1977). Loss of infectivity of brome mosaic virus RNA after chemical modification of the 3’ or 5’ terminus. Proc. Nat. Acad. Sci. USA 74, 2682-2686.
LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-685. LANE, L. C., and KAESBERG, P. (1973). Multiple genetic components in bromegraas mosaic virus. Nature
New Biol.
232, 40-43.
LAPinor, Y., DE GROOT,N., and FRY-SHAFRIR, I. (1967). Modified aminoacyl-tRNA. II. A general method for the preparation of acylaminoacyl-tRNA. Biochim. Biophys. Acta 145,292-299. MCLEESTER, R. C., and HALL, T. C. (1977). Simplification of amino acid incorporation and other assays using filter paper techniques. Anal. Biochem. 79, 627-630. PINCK, L., and HIRTH, L. (1972). The replicative RNA and the viral RNA synthesis rate in tobacco in-
EMV RNA AMINOACYLATION fected with alfalfa mosaic virus. Virology 49, 413-425. PINCK, M., GHAN, S. K., GENEVAUX, M., HIRTH, L., and DURANTON, H. (1972). Valine specific tRNAlike structure in RNAs of two viruses of turnip yellow mosaic virus group. Biochimie 54, 10931094. PINCK, M., GENEVAUX, M., and DURANTON, H. (1974). Studies on the amino acid acceptor activities of the eggplant mosaic viral RNA and its satellite RNA. Biochimie 56, 423-428. PINCK, M., and HALL, T. C. (1978). Aminoacylation properties of eggplant mosaic virus RNA: Separation and association of tRNAs. Virology 88, 281-285. PLEIJ, C. W. A., MELLEMA, J. R., and BOSCH, L. (1977). The occurrence of the coat protein messenger RNA in the minor components of turnip yellow mosaic virus. FEBS Lett. 80, 19-22. PLEIJ, C. W. A., NEELEMAN, A., VAN VLOTENDOTING, L., and BOSCH, L. (1976). Translation of turnip yellow mosaic virus RNA in vitro: a closed
AND TRANSLATION
365
and an open coat protein cistron. Proc. Nat. Acad. Sci. USA 73,4437-4441. SHIH, D. S., and KAESBERG, P. (1973). Translation of brome mosaic viral RNA in a cell-free system derived from wheat embryo. Proc. Nat. Acad. Sci. USA 70, 1799-1803. SHIH, D. S., and KAESBERG,P. (1976). Translation of RNAs of brome mosaic virus: The monocistronic nature of RNA1 and RNA2. J. Mol. Biol. 103, 7788. STUDIER, F. W. (1973). Analysis of bacteriophage T7 early RNAs and proteins on slab gels. J. Mol. Biol. 79, 237-248. SZYBIAK, U., BOULEY, J. P., and FRITSCH, C. (1978). Evidence for the existence of a coat protein messenger RNA associated with the top component of each of three tymoviruses. Nucleic Acids Res. 5, 1821-1831. WEBER, K., and OSBORN,M. (1969). The reliability of molecular weight determinations by dodecyl sulfate polyacrylamide gel electrophoresis. J. Biol. Chem. 244, 4406-4412.